This NSF fellowship program focuses on the intersection of AI and biological sciences to strengthen biotechnology innovations. Fellows develop expertise in both AI methodologies and experimental biology through two years of independent, fellow-driven research. As your sponsor, we can provide the research environment, mentorship, and resources to support your vision.

Research Alignment

Proposals that would align well with our lab’s expertise involve using AI to understand how to culture currently uncultured microbial lineages. This could include extremophiles with biosynthetic or bioengineering capacity that transform waste streams into consumer or industrial products, operate at environmental extremes, or contribute to the circular economy in other innovative ways.

We’re interested in sponsoring candidates who have experience using and training AI tools to screen existing databases (genomic, physiologic, metabolic) and who can design new experiments to inform and refine AI models. The approach we envision creates feedback loops: AI predictions guide cultivation strategies, experimental results validate and improve the models, and iterations accelerate discovery of microbes we can’t currently grow but whose genomic potential suggests biotechnological applications.

What We Offer as Your Sponsor

Research Infrastructure: The Arizona Culture Collection provides thousands of bacterial isolates as potential training datasets and validation resources. Our lab maintains both anaerobic and aerobic cultivation capabilities, including high-throughput flow cytometry. Through University of Arizona core facilities, you’ll have access to cell sorting, PacBio, Nanopore, and Illumina sequencing, and comprehensive microscopy facilities. These resources provide experimental capacity to test AI-driven predictions and generate high-quality training data.

Share

Structured Mentorship: We use Individual Development Plans (IDPs) to provide individualized mentorship focused on your career vision. Through biannual meetings, we create space to step back from day-to-day work and ensure you’re making progress toward your bigger picture goals. This isn’t performance evaluation—it’s about understanding your vision for your career and helping you achieve it. You’ll have opportunities to mentor undergraduate researchers, developing your own mentorship voice while accomplishing your project goals.

Collaborative Environment: You’ll be part of the BIO5 Institute community and broader University of Arizona research ecosystem, with connections to computational scientists, biotechnology researchers, and potential collaborators. The university offers extensive professional development through Postdoctoral Affairs and the Postdoctoral Association. If you move into an academic position after your fellowship, the microbes, resources, and techniques you develop can go with you to seed your independent career—no strings attached.

Lab Culture: We’re committed to inclusive excellence and supporting diverse perspectives in science. We value productive self-starters who take action. People with clear opinions who can commit to a direction even with disagreement, and scientists who ensure everyone is heard. We expect 40 hours of work per week with a mindful ramp-up period, direct feedback, and a team that shows up for each other. We want you to have a sustainable, healthy relationship with your work.

About You

You’ll have experience with machine learning approaches, database mining, experimental design optimization, or related AI technologies. You’ll bring your own vision for how to apply these tools to fundamental questions in microbial cultivation with biotechnological impact. You’ll have defended your doctoral degree before the fellowship start date and demonstrate strong communication skills and commitment to independent problem-solving.

Important Notes on Eligibility

Please note that there may be limitations in the NSF PRFB program for international candidates. Applicants must be a U.S. citizen, U.S. national, or U.S. permanent resident, i.e., have a “green card,” at the time of submission. You’ll need to evaluate your eligibility independently by reviewing the program guidelines at the link below.

Next Steps

If you’re interested in developing a proposal for this opportunity:

1. review the program guidelines

2. email me (paulcarini (at) arizona . edu) a 200-400 word summary of your proposed work. This should outline your vision, how AI and experimental microbiology would intersect in your approach, and how our lab specifically can help you accomplish your vision.

After reviewing your summary, we’ll evaluate whether your proposed research aligns with our program expertise and reply to let you know if I can serve as your sponsor with brief feedback. Remember—this is your fellowship and your research program. We’re here to provide the environment, mentorship, and support to help you succeed.

Contact: Paul Carini (paulcarini (at) arizona . edu)

The Sonoran Desert taught me something important during the pandemic: life here doesn’t just survive dryness—it waits.

Walking with my family through the desert during those long months of lockdown, I’d watch the seemingly lifeless soil. 2020 was a particularly dry year: no rain for months. Temperatures pushing past 110°F. The ground baking under relentless sun. But I knew the soil microbes beneath our feet were doing something remarkable while waiting for water. I just wasn’t sure what.

This waiting became the question that drove our research for the next five years.

Subscribe now

A Project That Refused to Die

Ryan Bartelme started this work in 2019, before any of us had heard of COVID-19. We had just isolated an Arthrobacter from subsurface forest soil in the Santa Catalina Mountains. The strain grew reliably, was easy to work with, and came from the kind of periodically dry soil we wanted to understand. It seemed like the perfect system to ask: What actually happens inside a soil bacterium when it dries out and then gets wet again?

The questions were straightforward. Most desiccation studies focused on rapid drying, flash-drying cells to see how they responded to sudden stress. But soil doesn’t dry overnight. We wanted to study slow drying, with proper hydrated controls so we could be certain the changes we saw were specifically due to desiccation. And while plenty of papers studied desiccation, fewer investigated rehydration: the journey back to life.

Then the pandemic hit.

The lab shut down in March 2020. When we reopened, it was with reduced schedules, social distancing, and a rotating cast of lab members as people graduated or had their bandwidth consumed by pandemic chaos. The project survived three years of disrupted science, passed between hands like a relay baton.

Ryan designed the humidity control system using saturated salt solutions to precisely control relative humidity. The approach was elegantly simple: different salts create predictable humidity levels, allowing us to slowly walk cells from 100% humidity down to 26% over two weeks, mimicking gradual water loss in real soil. The experimental design incorporated three essential features: hydrated controls running in parallel, slow desiccation over 14 days rather than rapid drying, and (honestly, this surprised us when it worked) rehydration with water vapor alone. No liquid water. Just shifting the dried cells back to 100% humidity and waiting.

Frozen in Carbonite

The first hint that something unusual was happening came from Izzy Viney’s Master’s work analyzing intracellular metabolites. She noticed RNA breakdown products accumulated in desiccated cells. This puzzled us. If RNA was degrading, there should be less intact RNA in dried cells.

But we could extract RNA. Not just ribosomal RNA, but messenger RNA too. Enough to sequence and analyze gene expression patterns.

This didn’t make sense. RNA is supposed to be transient. In actively growing bacteria, most mRNAs have half-lives measured in minutes. Environmental microbiologists use RNA presence as a marker of activity precisely because it turns over quickly in growing cells.

Yet here we were, extracting intact RNA from cells that had been desiccated for several weeks.

The full picture only emerged once Christina Guerrero completed the transcriptome analysis, Melanie Kridler finished the culture work and qPCRs, and Adriana Gomez-Buckley confirmed cells weren’t forming spores.

The pattern was striking. Gene expression changed dramatically as cells dried: mounting stress responses, producing protective compounds. Then, once cells were desiccated by day 8, transcription essentially froze. Between days 8 and 14, there was no significant change in RNA composition. The transcriptome was static.

The same thing happened with metabolites. Dramatic restructuring during drying. Then stasis during full desiccation. Then another wave of change upon rehydration.

Like Han Solo frozen in carbonite, the cells had pressed pause. Their molecular machinery preserved, waiting.

Water Vapor Whispers

The revelation came with rehydration. When we exposed dried cells to water vapor—just humidity, no liquid water—they woke up. Hundreds of genes increased in abundance. Energy mobilization. DNA repair. Stress response machinery.

Water vapor alone was sufficient to trigger resuscitation.

This finding excites me because it matters for real-world microbial ecology. Desert microbes don’t necessarily need liquid water to wake up. High humidity events or fog deposition might trigger resuscitation through vapor-phase rehydration. This has implications for understanding microbial activity in fog deserts like the Namib, in biological soil crusts, and anywhere vapor-phase water transfer occurs before liquid water becomes available.

What This Changes

This work challenges how we interpret molecular data from dry environments. Molecular ecologists assume RNA indicates activity. RNA:DNA ratios supposedly separate active from inactive populations. Metatranscriptomics presumably captures what communities are actively doing. But our data showed that RNA—both ribosomal and messenger—remains intact and extractable from cells that are metabolically dormant for weeks.

Soil rarely maintains constant moisture. In dryland ecosystems—covering about 30-40% of Earth’s land surface—drying is the norm. If you extract RNA from field soil, you’re inevitably capturing a mixture: some from active cells in wet microsites, some from dormant cells preserving transcripts from the last time they were hydrated.

When you grind up soil and extract RNA, you’re averaging across millions of cells in different physiological states. The transcripts you sequence might reflect what cells are doing now... or what they were doing the last time water was available. We can’t currently separate these populations.

The nucleotide breakdown products that first caught Izzy’s attention tell their own story. They accumulate during the transition to dormancy and persist through desiccation, then return to baseline upon rehydration. We think they represent internal RNA recycling. Components accumulate because there’s no new synthesis to use them. When water vapor triggers resuscitation, cells likely draw on this internal pool to rebuild RNA and restart metabolism.

A Pandemic-Era Science Story

Looking back, this paper is as much about persistence as the bacteria it studies. Ryan designing the experimental system. Izzy taking over when Ryan left, building her Master’s thesis and NSF Graduate Research Fellowship around these questions. Melanie’s patient culture work and RNA extractions. Christina’s transcriptome analysis. Adriana’s work ruling out spore formation.

The science emerged from walks in the desert during a global pandemic, from discussions about what questions actually matter, and from determination to understand not just whether bacteria survive desiccation but what they’re doing while they wait.

Soil bacteria hit pause when water disappears. They don’t just endure—they maintain themselves in a state of suspended animation, preserving their molecular machinery until conditions improve. Then water vapor whispers that things are changing, and they begin the careful work of waking up.

That’s worth understanding. Not just because it challenges how we interpret RNA in dry soils, but because it reveals something fundamental about microbial resilience in a world where drought is becoming more common. These bacteria have been waiting out dry spells for millions of years. Maybe there’s something we can learn from their patience.

The work was supported by the National Science Foundation and the Joint Genome Institute.

If you enjoyed this post, you might also like our previous “behind the paper” stories about relic DNA in soil, compact microplate readers, and genome evolution in deep-sea bacteria.

We can't.

The traditional approaches to research integrity—honor systems and detection tools—are broken in the age of generative AI.

The AI Spectrum

AI use in research isn't a simple case of "cheating" versus "legitimate use." It exists on a spectrum that hasn’t been adequately mapped. At the most problematic end, we have blatant academic dishonesty: submitting entirely AI-written papers without disclosure, fabricating data or hallucinating citations, creating peer reviews. These cases are clear violations that most would universally condemn.

At the acceptable end, we find uses that most would consider legitimate: AI-based citation formatting, coding and debugging analyses, brainstorming, basic editing and proofreading. Even here, though, questions remain about what level of disclosure is appropriate.

But, like many things in academia, the real challenge lies in the massive gray area where most people operate. This includes using AI to restructure your literature review, having AI help with English language editing, letting AI suggest analytical approaches for your data, using AI to generate multiple versions of your abstract, or having AI help respond to reviewer comments. Some of this feels reasonable. Some feels questionable. All of it depends on context, disclosure, and norms that don't exist yet. The gray area is huge, growing, and context-dependent. What's acceptable varies by discipline, institution, journal, and even across researchers.

The Honor System Is Dead

Academic research has always relied on trust. We trusted that researchers accurately reported methods, didn't fabricate data, and properly attributed sources. This worked when misconduct required deliberate, technically complex deception. AI changes everything.

For the first time, research assistance is available to anyone with an internet connection. An undergraduate can generate a literature review draft in minutes that used to take a doctoral student months to write. The barrier to what we might call "research enhancement" (or misconduct, depending on your perspective) has dropped to nearly zero.

More fundamentally, the honor system assumes clear boundaries. But AI use exists on that messy spectrum where boundaries are fuzzy, contextual, and shifting. When a tool can help with everything from writing to analysis to idea generation, asking researchers to self-police becomes impossible. The honor system also relies on peer oversight. But AI-assisted work often looks indistinguishable from human-generated content. Peer reviewers, already overwhelmed and time-limited, can't realistically evaluate every submission for appropriate AI use. And we already know many reviewers are using AI tools themselves.

Although the lines are already blurry, AI companies are developing systems to further obscure them until they disappear. The next generation of AI tools won't just help with writing—they'll integrate seamlessly into research, offering real-time suggestions, automated analysis, and assistance at each step. The goal is to create tools so sophisticated and produce content so human-like that the distinction between human and AI contributions becomes zero. We're not just facing a current problem with fuzzy boundaries—we're heading toward a future where those boundaries won't exist.

AI Detection is Not the Answer

Faced with these limitations, we’re turning to AI detection tools. These promise to identify AI-generated text and restore some objective measure of authenticity.

This approach is doomed.

First, as many of us know, the tools don't work. AI detectors produce false positives, flagging human writing as AI-generated, and false negatives that miss obvious AI content. As AI systems improve, detection becomes even harder as detection will always lag behind generation technology. A major reason detection tools don’t work is they can't handle nuance. They might catch a fully AI-written paragraph, but what about text that's been AI-edited, AI-restructured, or AI-inspired? What about ideas generated through AI brainstorming? The sophisticated ways researchers use AI are invisible to detection algorithms.

Further, detection creates a surveillance culture. When we’re focused on catching violations rather than fostering responsible use, AI use is driven underground. Instead of encouraging thoughtful integration, we get secretive use and defensive researchers. Most importantly, detection misses the point. The goal shouldn't be eliminating AI from research. It should be to ensure AI use enhances rather than undermines research integrity.

What Actually Works

If honor systems and detection tools aren't the answer, what is? We need a fundamental shift from policing AI use to normalizing transparency about it.

We need to normalize disclosure. Researchers should document and disclose AI workflows just like any other methodological choice. This isn't confession—it's reproducibility¹. When someone uses AI to help analyze thousands of documents, that's methodology worth reporting. Discipline-specific guidelines would be a good step toward normalizing disclosure. A historian using AI to process archival materials has different integrity considerations than a microbiologist using AI to help write a methods section.

Focusing on education should be a priority. Researchers need training on AI capabilities, limitations, ethical implications (including environmental²), and best practices. Most people want to do the right thing—they just don't know what that is. Instead of trying to detect AI use after the fact, build submission systems that encourage disclosure. Make it easy to report AI assistance. Create templates for AI methodology sections.

We also need boundaries around human connection. Although AI is useful in many aspects of research, it shouldn't replace collaboration and mentorship. Using AI to draft emails to colleagues, generate responses in Slack or Basecamp, or simulate important face-to-face discussions undermines the relationships that make science meaningful and productive. The conversations, spontaneous remarks, and trust-building that happen through direct human interaction can't be automated away. AI should augment our capacity to do research, not substitute for the human connections that drive the science forward.

Let Me Be Clear: I Use the Tools

Writing is the part of my job I love the most! Yet, I've found AI has enhanced both my writing and my coding and just about anything else I use it for. Yes, there are hallucinations. But I'm the expert and I still maintain control. So, it's up to me (as it should be) to verify everything produced. And I often find errors! Yet, I struggle to see the downside of using these tools to improve the communication and pace of analysis of scientific data. Who wants to read a crappy paper? And don't try to tell me that no one has ever mis-coded something and come to the wrong conclusion. Mistakes happen in coding with humans too. The difference is that with AI assistance, I can iterate faster, explore more approaches, and communicate more clearly. The responsibility for accuracy, interpretation, and scientific validity remains entirely mine. That's how it should be.

Speaking of transparency, I used AI assistance in writing this post. I used AI tools to help brainstorm ideas, consider different perspectives, and improve clarity in several sections. I also used AI to help me decide what was superflous and condense repetition in previous drafts. The core ideas, arguments, experiences, and overall structure are my own, but AI helped me articulate these ideas more effectively. This is exactly the kind of disclosure that should become routine in academic work. (Nota bene: I was using the em dash long before AI adopted it.)

The Stakes

The choices we make now will shape scientific knowledge production for decades. If we treat AI as either forbidden, we'll drive its use underground. If we ignore it, we miss opportunities to harness its potential to actually do good. The path forward requires prioritizing transparency and recognizing that research integrity in the AI age will look different from what came before. Researchers are already using AI. The question isn't whether this will happen—it's whether we'll create systems that encourage responsible use or drive it into the shadows.

Share

¹The nature of generative AI means that identical prompts can produce different outputs, creating reproducibility challenges that traditional research methods don't face.

²AI tools come with significant environmental costs through their energy consumption and carbon footprint. Each query to large language models requires substantial computational resources. As we advocate for responsible AI use in research, we should also consider these environmental impacts as part of our ethical framework.

Dear New PhDs,

Stop me if you've heard this one: "We're interested in hosting you as a postdoc, but you'd have to bring your own funding."

Or maybe it's the folks suggesting your analytical skills are being superseded by AI.

Perhaps it's both—a perfect storm where funding has dried up just as AI threatens to reshape the scientific landscape.

If you're graduating into this reality, you're facing a different landscape than the one I did a decade ago.

First: this absolutely sucks.

Many of you started your PhD just before or during the COVID pandemic. Lab shutdowns. Remote work. Disrupted research. Now as you complete your degrees, you're facing an unprecedented funding future. You've been screwed on the way in and on the way out.

As a PI who sees your talent, your dedication, and your potential, I get the unfairness of this. You've done everything right. The system is failing you, not the other way around.

I'm sure you feel unseen right now. Like your years of work don't matter. Like the expertise you've built has no place. But we do see you! Your mentors, the faculty, the researchers—we see your capability and your contribution. This institutional failure doesn't diminish what you've accomplished or what you're capable of!

The emotional toll is real. It's not just about money (though money matters). It's about watching a dream career crumble. It’s about feeling like the future you trained for vanished. It’s about wondering if you wasted years of your life. These feelings are valid. Make no mistake: this is grief. For the career path that was supposed to be, for the predictability you thought you'd have, for the sense that hard work would be rewarded.

What now? A few practical thoughts from someone still figuring it out alongside you.

This moment is training

This moment isn't something to just survive. It's training in the qualities that will define the next generation of scientific leaders.

You're being forced to develop resourcefulness. Creativity. Resilience under pressure. While your predecessors learned to navigate established systems with relatively ample funding, you're learning to build new ones.

Every constraint you're working within—limited funding, uncertain career paths, rapidly changing technology—is teaching you. As you emerge from this, you will be uniquely equipped to thrive in uncertainty. To build value from limited resources. To find innovative solutions.

This isn't a consolation prize. This is recognizing that adversity, when met with intention, becomes the thing that prepares you for unknown future challenges.

Focus on what's within your control

When systems collapse, the natural response is panic or paralysis. Resist both.

Your ability to think systematically, solve complex problems, and learn rapidly remain valuable regardless of funding structures.

What you can control is your skills pipeline. Keep building capabilities that transcend funding models. Your network matters now more than ever! Connect with folks whose work you admire, not just those with funding. Work on your adaptability threshold by practicing getting comfortable with uncertainty. Maintain your daily work habits.

Stay focused on the present. Energy spent dwelling on the broken system, COVID disruptions, and the unfairness of it all drains you. Worrying about future scenarios creates anxiety that undermines your ability to act effectively now. You only ever have 'now' to make progress, ask good questions, or build something. Guard that focus carefully.

What you can't control: Federal mandates. Congressional appropriations committees. University administrative decisions. Whether funders decide basic research matters again. The opinions of people who think you should "get a real job."

Put your energy exclusively in the first category.

Don't try to game the system

I've watched colleagues transform into professional grant-chasers, becoming more skilled at working funding angles than conducting actual research. They've built careers around predicting what buzzwords program officers want to hear. With federal funding diminishing, and with a completely reworked set of research priorities, those specialized gaming skills are depreciating assets. The next system will have different rules, different gatekeepers, different priorities.

Instead, build your reputation on substance. Do good work that matters to you and can be explained to others. The format and funding will change. The value of your genuine contribution won't. Whether your research is funded through federal grants, industry partnerships, or models that don't yet exist, the quality of your thinking and your approach will always matter.

Don't go dormant

Equally dangerous is the "wait and see" approach. There might be a tendency to freeze in place, thinking the odds are too long. I'm guilty of this myself! I tell our lab group to keep moving forward and do good work, because it matters now mroe than ever. Unlike dormancy in microbes, hibernation is rarely a successful professional strategy.

Remaining scientifically alive means continuing to ask questions that matter. It means designing ways to answer them, even with limited resources. It means sharing what you learn through whatever channels remain, and engaging with others' ideas. Find ways to stay engaged, even if it looks different than you imagined.

Creativity will be your most valuable research tool. Look for problems that can be tackled with ingenuity rather than resources. Collaborate to share costs and expertise. Make use of open-source tools, citizen science, and creative partnerships with industry or community organizations. Some of the best scientific insights have come from folks working with constraints.

Expand your vision

Scientists are increasingly exploring areas that don't feel central to their scientific identity at first. Intellectual property law. Project management. Business development. Data visualization. These adjacent skills expand what you can recognize as opportunity and create new pathways for applying your scientific training.

Consider exploring management and leadership. Leading teams effectively is rare, yet always valuable. Business fundamentals matter—learning how resources flow and decisions get made isn't "giving up." Communication across disciplines creates value when you can be the translator between technical and non-technical worlds. Data science and visualization skills will only become more essential as we make meaning from complexity. Understanding regulatory and policy landscapes provides both insight and opportunity.

The goal isn't to abandon science but to create more surface area for work to stick to.

Find a job

Let's be honest: none of this advice pays rent. You need income. You need health insurance. You need to eat. The financial pressure is real and immediate.

So find a job doing something. Anything that uses your analytical skills while you figure out the next move.

Don't view that job as giving up. Every experience builds capability for what comes next. And it may open doors that you didn’t know existed.

That consulting role teaching you how businesses make decisions? You're learning how organizations function. The data science position showing you how to communicate complex information to non-experts? You're developing translation skills that will serve you everywhere. The regulatory job revealing how policy gets made? You're understanding the system that shapes research priorities. Even the completely unrelated work that pays the bills while you volunteer on research projects nights and weekends? You're building resilience and time management skills.

Your PhD didn't expire the moment you couldn't find a postdoctoral position. It's portable. It's valuable. And sometimes the most interesting scientific careers are built by people who took detours through other worlds first.

The goal is to recognize that each experience—if approached with intention—expands what you're capable of doing next.

Consider strategic education (maybe)

More education isn't automatically the answer. However, targeted learning can create leverage. A colleague recently completed a one-year master's in GIS technology that transformed her understanding of how to communicate science with maps.

Ask yourself: Does this additional education create new capabilities or just delay decisions? Can you learn this through means other than formal education? What's the actual return, not in dollars, but in options created? Will this help you create value regardless of funding models?

Share

No one knows what happens next

The most honest thing I can tell you: no one—not a mid-career faculty like myself, not industry leaders, not policy experts—truly knows what the hell is going on or how science will reorganize with reduced federal funding. We're all thinking on our feet. Those of us in PI positions see your struggle and wish we had clearer answers or more secure pathways to offer you.

This uncertainty is frightening. But it's also where possibility lives.

Without established pathways, you're free (and now required) to create your own. Each generation of scientists has faced what felt like existential challenges. Scientific funding has evolved—from individual patronage to systematic government support to today's diversified landscape, which is now giving way to something else entirely. Science has survived religious control, wars, economic collapse, and radical technological shifts. This current crisis will eventually seem like just another transition point in the long arc of human knowledge.

What matters is that you keep asking important questions. Find ways to pursue them. The "how" will remain fluid, but the "why" is in your hands.

The researchers who emerge from this period will be the architects of whatever comes next. Your generation is being forced to reimagine how scientific knowledge gets created, funded, and shared.

That's not just adaptation. That's leadership.

With solidarity and hope,

Subscribe now

In February, I caught what seemed like a run-of-the-mill non-COVID viral illness. I recovered from the acute infection, but apparently my immune system didn't receive the "stand down" signal. Instead, it's been attacking my own body, resulting in substantial pain that has persisted since.

At first, I dismissed it as excessive muscle soreness from pushing too hard in the gym (not unusual; humble brag). But this wasn't my normal post-workout pain. It never went away. By six weeks in, the situation deteriorated substantially—I couldn't sleep through the night, put on a shirt without wincing, get up from a chair or out of a car without assistance, or manage to get out of bed without an orchestrated sequence of movements and a bunch of pain. My kid mocked my “old man joints.”

Subscribe now

The journey to figure out what's happening has been frustrating and remains incomplete. And even my doctors don’t fully understand what's going on. My previous baseline slipped away quickly—I went from deadlifting and squatting reasonable amounts of weight to being unable to perform even a single pushup within a few weeks. Over-the-counter medications were ineffective.

I'm now receiving treatment, but the course of therapy and long-term outcomes are unclear. It’s an odd limbo when you're being treated for something that isn't fully understood—a mix of hope, uncertainty, and the practical reality of managing the day-to-day while waiting to see if the treatments will work.

I'm sharing this experience for a few reasons:

1. The chronic pain knowledge gap. When people who don't experience chronic pain hear about someone with chronic pain, they genuinely have no concept of what it means. And because of this disconnect, chronic pain is dismissed. I am guilty of this myself. Chronic pain is all-encompassing in a way that's difficult to articulate. It hurts to sit. It hurts to stand. It hurts to lie down. When those three activities cover essentially everything a body can do, you're left with few options for relief.

   Chronic pain isn't just physical—it affects your mood, your relationships, your ability to work, and your basic sense of self. We could all benefit from being more compassionate when thinking about these “invisible” disorders and the holistic toll they take on people's lives.

2. The viral immunity black box. My immediate response as I started to piece things together was to go to the literature. My conclusion: There is so much we still don't understand about how viruses interact with human health and physiology. My personal records (more below on this) strongly point to a viral trigger. But fundamental questions remain: Why did this virus trigger this response? What virus was it? What's the mechanism by which it's affecting my body, and how common are these kinds of post-viral immune dysregulations?

   Long COVID is now infamous. Millions worldwide have experienced persistent symptoms after a COVID infection, ranging from fatigue and brain fog to cardiovascular issues and chronic pain. My situation is a microcosm of what we're seeing at a population level—viruses are thought of as temporary ailments, but can have long-term effects. Whether it's long COVID, chronic fatigue syndrome following EBV, or my own post-viral condition, the reality is that our understanding of how viruses can rewire our immune systems remains frustratingly incomplete.

   As such, this experience has reduced my tolerance for anti-science positions to absolute zero (not that they were much above zero). When I hear people dismissing the importance of vaccine research, defunding virus research, or spreading misinformation about public health measures, I now have a visceral reaction. And yes, I'll likely be masking more frequently in the future. The cost-benefit calculation looks different now.

3. The value of contemporaneous health records. I’m a bit of a data nerd. I've kept meticulous personal health records for years—essentially a lab notebooks full of data about my body. Near-daily entries note what hurts and when, workout details, heart rate data, blood pressure readings, weight, body measurements, and any other metrics I can easily track at home without specialized equipment.

   These detailed records proved invaluable in connecting the viral infection to the onset of symptoms, establishing timelines, and identifying patterns that might otherwise have been missed. Without this data, diagnosing the underlying cause would have been nearly impossible. I admit, it’s a bit compulsive. But it makes spotting the deviations from baseline much easier.

   

4. The medical system paradox. The medical system is simultaneously miraculous and broken. The diagnostic capabilities, treatments, and expertise available are truly remarkable. Yet, navigating online portals, scheduling appointments, dealing with insurance requirements, managing copays, and meeting complex referral prerequisites for tests can feel like a part-time job in itself.

   I'm fortunate to have an excellent care team and support at home, but I genuinely cannot fathom how elderly patients or those with cognitive impairments manage to navigate this system. The barriers to getting necessary care are substantial, even for someone with resources, education, and privilege.

   Join the quorum already reading UnCultured! Subscribe for free to receive new posts.

I don't have tidy conclusions or a happy ending (yet!). I'm still in the middle of this journey, still figuring out what works and what doesn't, still trying to rebuild strength by doing what I can to find a new normal.

What I do know is that this experience has changed how I view health, pain, medical research, and our complex relationship with viruses. And there’s a humility that comes with having your immune system decide, for reasons unknown, to turn against your own tissues.

It's a reminder that for all our scientific advances, there are areas of human biology that we're only beginning to understand. And it's a call for more empathy toward those whose bodies are fighting battles we can't see.

These expectations can be conveyed easily through an “expectations document.” Below, I’ll explain what they are, why they’re important, and how to construct your own. I’ll also share our most recent iteration of our internal document.

What is an Expectations Document? It’s exactly what it sounds like: A written agreement that outlines the responsibilities, commitments, and goals of both mentors and mentees. Unlike informal conversations which can be forgotten or misinterpreted, a written document provides clarity and consistency that benefits everyone involved.

These documents aren't one-sided lists of demands from the PI. Rather, they're mutual agreements that define what mentees can expect from their mentors and what mentors need from their mentees. They create a scaffold for communication, feedback, and growth while establishing a lab culture built on respect and transparency. I’ll typically go over these documents during the first week of employment in our lab.

Why Expectations Documents Matter. For mentees, expectations documents provide clear guidelines about performance standards, communication protocols, and participation in lab activities. They reduce anxiety by eliminating guesswork about what "expected" looks like in the lab context. We also post these on our lab website so prospective lab members gain insight into the lab's values and culture before joining. This helps them make informed decisions about their future scientific home. Perhaps most importantly, written expectations ensure consistent standards across all lab members, regardless of background or experience level.

I directly and indirectly benefit equally from these documents. They help guide healthy boundaries and work patterns from the beginning of the relationship. Common misunderstandings that might otherwise lead to conflict can be preemptively addressed. The document serves as a concrete tool for accountability and performance discussions, taking some of the emotion out of potentially difficult conversations. When challenges arise, both parties have a reference point to guide their discussions.

Our internal Expectations Document. Our expectations document reflects our commitment to intentional mentee training. We emphasize creating an environment where people from any background can thrive while conducting rigorous, innovative research in microbiology. Our current Lab Expectations can be found here. Feel free to copy, use or reinterpret any of these details for your own use. Several portions of this were commandeered from others (though I can’t recall exactly who).

Creating Your Own Expectations Document. If you're considering creating an expectations document for your lab or mentoring relationship, several key elements deserve attention. Start with your core values—the foundational principles that guide your work together. For us, that includes laboratory and emotional safety, a welcoming work environment for everyone, communication expectations, integrity expectations (including the use of AI), meeting frequency and protocols for questions or concerns. Define performance standards for each lab member rank clearly so everyone understands how progress and success are measured at their career stage.

Work environment considerations should include schedules, attendance expectations, and physical presence requirements. Consider adding authorship and recognition policies for publications and acknowledgments (here are ours). Outline both opportunities and responsibilities for professional development, possibly through the use of individual development plans. Finally, establish processes for addressing disagreements or concerns before they arise.

An expectations document isn't effective if it just sits in a drawer or folder. For maximum impact, review it regularly with all lab members. Treat it as a living document that evolves with your lab's needs and experiences. I revisit our document roughly annually and revise as needed. Recently, we added a section dedicated to our AI use guidelines. We reference it during onboarding, evaluations, and when addressing concerns. Most importantly, use it as a tool for reflection and growth, not just enforcement of rules.

Share

It's a bit of work at the beginning, but so worth it. The time invested in creating clear expectations pays dividends in reduced conflict, increased productivity, and more meaningful mentor-mentee relationships. By articulating our expectations and commitments, we create space for both structure and creativity—the ideal environment for scientific discovery and professional growth.

Your most important mentors are the shadow mentors.

You know what “shadow mentors” are, even if you don’t recognize the term. The professor from a different university who gave you timely career-saving advice at a conference. The scientist who DM'd you after your terrible first talk to say "been there, here's what worked for me." The senior postdoc from another lab who showed you how to troubleshoot that assay when your supervisor was unreachable.

And let's not forget the social media contact you started messaging with after they left a comment on your method-troubleshooting thread. The one who's offered virtual support through three job changes, two rejected manuscripts, and that time you almost quit science. The one who somehow always responds within hours, despite living in a completely different time zone.

Then there's the person who reviewed your fellowship application, job materials, or grant proposal for no benefit to themselves. The one who spent hours marking up your CV, suggesting major revisions to your research statement, and helping you convert academic jargon into something the review committee might understand. The person who had no stake in your success but treated your application like it was their own career on the line.

They didn't have to do that. But they did.

Shadow mentors have no official responsibility for your success. There's no formal relationship. They don't benefit from your achievements. They just…help you.

And they're absolutely crucial to scientific careers.

Share

Why shadow mentors are different.

Official mentors are stuck evaluating you. Even the good ones can't fully escape this dynamic. They're invested in your projects. Your results—good or bad—reflect on them.

Shadow mentors don't have this baggage. They can tell you when your research question is fundamentally flawed without worrying about the grant they wrote around it. They can advise you to switch labs or fields or to pursue your non-science life dreams without feeling personally rejected.

They're not trying to mold you into their scientific image. They're just responding to you as a human.

You probably need more of them.

If your only mentorship comes from people with formal power over your career, you're missing out. Shadow mentors provide perspective from different subfields and institutions. They offer career advice without political complications. You get technical knowledge your mentors lack and connections to entirely different networks. Plus they might be someone safe to vent to about your actual mentor.

The best part? You can have several. Many! A cabinet of shadow mentors addressing different aspects of scientific life.

How to find Shadow Mentors?

Unlike formal mentorships, these relationships rarely start with anyone saying "I'd like to mentor you" or "Will you be my mentor?"

They begin with genuine connection. Ask a thoughtful question after a seminar. Follow up on a helpful conference conversation. Email them saying you loved their last paper (FYI, everyone LOVES this!). Engage with someone's work online. Be vulnerable and open about a challenge you're facing. Show appreciation when someone helps you.

The relationship grows organically from there. Sometimes it's just one pivotal conversation. Other times it evolves into years of periodic check-ins and camping trips with your families.

Being someone's shadow mentor.

You're probably already someone's shadow mentor without realizing it.

That early career researcher you chatted with after your talk? The student from another lab who asked how you got your analysis to work? The colleague who reached out about navigating a difficult workplace situation? You might be the exact right person at the exact right moment to provide perspective they can't get elsewhere.

Take these interactions seriously. Be honest about your experiences, especially the failures. Make introductions when you can. Follow up. You don't need to formally adopt anyone. Just be generous with your knowledge and perspective when the opportunity arises.

The shadow mentorship ecosystem.

Our scientific education systems are designed around hierarchical, formal mentoring. But shadow mentorships create a parallel network of support that's often more responsive, more honest, and more adaptable.

The best scientific communities cultivate both.

So reach out to that researcher whose approach you admire. Answer the email from the early career scientist with a thoughtful question. And thank the shadow mentors who've shaped your own path.

They probably have no idea how much they helped.

Share UnCultured

Soon, I found myself stalking colleagues' citation counts and ranking them against my own, turning my career into a secret competition. If someone’s h-index outpaced mine, I’d feel inadequacy creep in. If mine edged ahead, for a moment, I’d feel validation. This unhealthy comparison was amplified through social media's distorted lens—a showcase of others' achievements with failures filtered out. The perfect storm of confirmation and survivorship bias reinforced my most destructive insecurities.

These rituals were the first symptoms of how the "publish or perish" mentality had enmeshed with my sense of self. Other warning signs followed: rejections that felt like personal failures rather than statistically expected outcomes; the inability to be fully present with my family—instead, my mind racing about research; the compulsion to accept every project, committee, and opportunity until my calendar became a tetris grid with no empty spaces and spilling into the off hours.

The boundary between what I did professionally and who I was as a person had dissolved completely. I wasn't working as a scientist—I was engulfed by science with all the vulnerability that entails. When experiments failed or grant applications were rejected, it wasn't just a professional setback but an identity crisis.

As academics, we chronically undervalue our time. Without deliberately established boundaries, work seeps into every corner of life. Our culture celebrates endless productivity—more papers, more citations, more grants, more respect—creating an insatiable cycle of "more" that ultimately costs us both personal and professional well-being.

Share

As funding landscapes grow increasingly uncertain, examining the boundaries between self and career becomes vital. Many STEM academics pour decades into specialized research, allowing their professional and personal identities to merge completely—a fusion rarely acknowledged in academia. This raises crucial questions: What happens to your sense of self when experiments fail, grants disappear, or retirement arrives? What remains when being a scientist shifts from something you do to something you are?

In our research group, we intentionally try to break from these academic norms. We maintain that 40 hours weekly provides ample time for significant contributions—for graduate students, approximately 20 hours of coursework plus 20 hours of research. We deliberately promote—and share—disconnection outside work hours to foster well-rounded lives and sustainable careers. This philosophy starts with me. I rarely work weekends and remain unavailable after hours except for genuine emergencies. I've deliberately cultivated interests outside of academia*. Many center around quality family time and building relationships with people who either know nothing about microbiology or express only casual curiosity. These pursuits help me step completely out of my professional mindset.

Of course, certain periods inevitably require greater commitment—comprehensive exams for students, concurrent fellowship applications and time-sensitive experiments for staff or postdocs, and grant proposal deadlines for myself. However, these intense phases must remain exceptions rather than becoming the normalized standard that consumes our identity.

The journey from being consumed by my career to finding a more balanced outlook began with the simple yet powerful mantra: "Good enough is fine." This didn't happen overnight but evolved gradually as I incorporated practices that created a healthy separation between my professional and personal identities. By applying the following approaches, I began to disentangle my sense of self from my publication record and professional achievements.

1. Schedule non-negotiable personal time. Before my calendar fills with work-related activities, I block specific hours in my calendar for non-academic pursuits with the same commitment I give to my research and teaching. Whether it's a local sports league or a blacksmithing class, I treat these commitments as nonnegotiable. Taking care of myself first builds the foundation from which I can do my best work.

2. Establish clear boundaries with digital tools. I use dedicated devices or separate accounts for work and personal activities when feasible. I use apps, and strategic inconveniences, to block or avoid work-related emails and notifications during my personal hours. I extend this separation to social media, which subtly blurs the professional and personal boundaries (in addition to fostering anxiety about the state of the world). Finally, I enact intentional barriers against checking citation metrics, manuscript statuses, or academic social platforms during my off-hours—these quick "checks" often become extended work sessions that erode personal time.

3. Explore interests unrelated to your field. Deliberately choose activities that exercise different skills and mindsets than your academic work. If you spend your days analyzing data, try something creative like pottery or music. If you work in theoretical fields, try hands-on activities like gardening or woodworking. The cognitive contrast helps create distance from work identities.

4. Build relationships outside your department. Try to form friendships with people who aren't in your field or academia altogether. Join community organizations, recreational classes, or volunteer groups where you'll meet people with diverse professional backgrounds. These relationships provide perspective and conversations that don't revolve around complaining about the grind of publications, grant deadlines, or university drama.

5. Practice the "beginner's mindset." As academics, we're trained to be experts, which can make trying new things intimidating. Embrace being a novice at something—take an introductory class in a subject you know nothing about. The humility and curiosity required to learn something new counterbalances the expertise-focused academic mindset and reminds you that identity isn't tied to achievement.

My perspective transformed during the COVID-19 pandemic, forcing me to confront my academic identity fusion when the world paused, but my internal engine still raced at full speed. As we collectively navigated the uncertainty brought on by the pandemic, people showed remarkable understanding that life is complex and multifaceted. That brief window of understanding has given way to familiar pressures, but it revealed an important truth: our identities exist beyond our professional contributions. By intentionally creating space between who we are and what we do professionally, we build resilience against the inevitable ups and downs of academic life. The scientist in the lab and the person outside it can coexist without merging completely. Your research matters—but you matter more than your research.

Leave a comment

*Over the years, I've explored fermentation (admittedly still microbiology-adjacent), sausage making, fitness, automotive repair, leatherworking, blacksmithing, and woodworking—with welding next on my list. The communities surrounding these activities couldn't be more different from academia, providing balance and reconnecting me with my core identity beyond my profession. Some of these are captured in the images below.

I highly recommend subscribing to the Matters Microbial Podcast! I always learn something new when I watch or listen.

We talked about…

1. Me digging holes in my Wisconsin backyard as a kid, searching for what I thought were dinosaur bones but were just buried chicken bones from the neighboring tavern. And about how my academic journey wasn't smooth - I got kicked out of two universities before finding my footing - this early curiosity about what lies beneath the soil eventually led me to my current research.

2. My lab has built what we call the Arizona Culture Collection - about 3,000 microbial strains from various desert sites. We wanted to understand how non-spore forming bacteria cope with desiccation, so we developed specialized chambers to dry cells slowly over two weeks, mimicking natural conditions rather than the rapid drying typically used in labs. We found that bacteria can be revived using just water vapor - no direct contact with liquid water is needed. Perhaps our most surprising discovery was about RNA stability. Conventional wisdom says RNA degrades quickly, but we found that RNA profiles remain remarkably stable in dried cells for extended periods. This challenges our basic assumptions about using RNA as a marker for microbial activity in environmental samples.

3. We closed with a discussion of our diverse, talented researchers who bring different perspectives to this work. We have PhD students studying archaea and various aspects of bacterial survival.

The desert has taught us that microbes are more sophisticated than we often give them credit for. When you look at a dry desert soil, it might seem lifeless, but it's full of microorganisms that have evolved remarkable strategies to survive these harsh conditions. Understanding these strategies isn't just about satisfying scientific curiosity - it could have practical applications in fields ranging from medicine to agriculture.

Letters of recommendation are the currency in a futile cycle of professional evaluations.

Whether applying for graduate school, or a new job, there's a good chance the employer will ask for a letter of recommendation. The applicant secures a commitment to get recommendation letters from 3-5 referees. The referees write and submit their letters. The letter becomes a part of the applicant's application package.

The employer reads the 3-5 letters from each applicant. They use the letters to score or rank applicants.

Like a futile cycle, these letters are good for little more than dissipating energy. At best, they are a time drain. Damaging to the applicant at worst.

It's awkward for candidates to ask for letters of recommendation. Referees, sometimes swamped by requests, have limited time to write them. Employers or committees spend a ton of time reading them.

As a first-generation college student, this whole process was mysterious. I was a mechanic in a race car shop trying to break into academic circles. Professors intimidated me. How on Earth was I going to get a letter of recommendation? Who would I ask? Why would they spend time recommending me? What if they said no? What if the letter sucks? It took immense energy to simply ask for a letter. After I mustered the courage, and my referee agreed, this mysterious transaction occurred. They'd "write a letter" and submit it and I'd either get the position or not.

Now, I see the other side as an employer. I run a mid-sized research laboratory at a high-research activity university. I regularly review and write letters of recommendation for faculty, graduate students, postdocs, and staff. Letters range from short paragraphs to novellas in length. Some letters are glowing. Rarely are they critical. Most are in the middle between these extremes.

To the candidate, the entire process is invisible. Applicants rarely see their letters (unless they write them themselves, more on that in a minute). Search committees or employers rarely explain the role letters play in their evaluation.

There are several serious issues with using letters of recommendation to rank candidates. The letters that make it into application packets are subject to a huge selection bias. The applicant asks people for letters who are likely to write letters that are not bad. There is almost no chance that a letter makes it into a packet that is critical of an applicant. Thus, candid criticism can sink an application package, even if it's constructive.

All letters are anecdotal evidence. Everything in the letter can be true according to the referee's experience. And yet, it does not predict future success for the applicant. There is no guarantee that someone amazing in one situation will work well in the next. Likewise, someone with untapped potential in their previous position may excel in a new position. Context matters. There are almost infinite reasons for this—related to the job or otherwise.

Strong letters have exactly one thing in common: the referee's persuasiveness. The perceived strength of the letter is more about the referee's writing ability than the strength of the candidate. Good referees are masters of picking—and phrasing—which strengths to emphasize. And how to soften weaknesses. Yes, letters can help fill some of the between-the-lines aspects of the applicant. Yet, often what is not in the letter is as—or more—useful than what is in the letter. Thus, experienced evaluators may read between the lines of what is in the letter.

Share

Letters may promote gender and racial bias. For example, the gender or race of a candidate influences the length and descriptions contained within letters. These differences are subtle but affect the applicant's selection success. Such discrepancies may affect male applicants from underrepresented groups the most. (See here, here, and here)

There are power dynamics between referees and candidates. Some referees covertly or overtly use that power to coerce their understudies into performing more work. Likewise, candidates may join a group for the sole purpose of "getting a letter" for their next career step. They are not as interested in the work at hand as the utility of the group to provide a letter in the future (not to bash PreMed students, but this is common among them).

There is a lack of training on how to write letters. Few curricula include best practices for letter writing. What are the components of a letter of recommendation? How does one avoid gender or other bias? Some referees even ask candidates to write their own letters of recommendation. This is especially true for "high profile" referees who already have full plates. They may not know the candidate well. They may not have the time. They may not feel strongly about the candidate. Whatever the reason, they solicit a letter from the candidate, lightly edit it, and sign it. The evaluation committee is none the wiser.

Likewise, each employer or institution evaluates letters differently. These metrics are opaque to everyone but the folks reading the applications. Even within the group of people reading them, differences in interpretations may exist on the sentiments conveyed in the letters.

Recommendation letters are futile—a huge time burden for the referees and evaluators with very little return on investment. They are a relic from the 'good ol’ days' when a letter from the right person opened doors. They feign a level of selectivity and have become surrogates for evaluating the candidate's actual records.

If letters are unreliable, what can we use instead? It may sound obvious: focus on the materials prepared by the applicant. Read the personal statement or cover letter. Is their interest in the position genuine? Do they write clearly? (I'm a huge proponent of clarity of writing equating to clarity of thought.) Is the cover letter authentic? Does it have a life, or is it bland and flat? Do they explain discrepancies in their record (grades, gaps, other anomalies)? Is there evidence of reliability or leadership in their work history?

Some of you might be thinking "I enjoy writing letters for my students" or "I find letters to be useful!" I too enjoy writing letters for quality students. And recommendation letters may have utility in specific instances. For example, letters can contextualize reasons for poor grades. Or explain an applicant's reason for seeking a new position. But these things could and should be in the cover letter, in the applicant's own words. If not conveyed clearly by the referee, nuanced sentiments like these get lost in a sea of mediocre letters. Tired eyes writing letters and tired eyes reading them presents a lose-lose situation.

If recommendation letters are here to stay, perhaps we can re-envision them. The National Science Foundation's model for letters of collaboration might be a good place to start. NSF dictates the following text for letters of collaboration:

“If the proposal submitted by Dr. [insert the full name of the Principal Investigator] entitled [insert the proposal title] is selected for funding by NSF, it is my intent to collaborate and/or commit resources as detailed in the Project Description or the Facilities, Equipment and Other Resources section of the proposal.”

Similar verbiage could be adapted for letters of recommendation. The applicant uploads their cover letter or personal statement. In the statement, they explain their relationship with a referee. And reflect on how the referee perceives them as a candidate. Submission triggers an email to the referees who check a box next to this statement

“The personal statement submitted by [insert the full name of the applicant] for [insert the title of the position] accurately conveys my relationship to the candidate, and I agree they have the potential to succeed in the position described.”

And that's it! It's about as useful as a full letter to the application packet. Yet, it takes a fraction of the time for both the referee and the review committee.

In the meantime, we can't get around the fact that letters are a current obligation. And I’ll keep writing them because I want my students to get new exciting positions! Maybe I'll start including some variant of the following early on in the letter:

I recommend {name of candidate} for {name of position}. But my experience with them is an anecdote. There are a myriad of ways and reasons that you might not have the same experience (better or worse). Their skills are good enough to get them in the door, for the reasons I explain below. But how they perform from here on out will be up to you and how you support them.

Hopefully, recommendation letters will disappear from our ever-expanding analytics-based approaches to admissions and awards. And we can dedicate the energy spent writing letters to other aspects of our work and lives.

*Futile cycles are now known to play important roles in regulating metabolite concentrations.

Share UnCultured

In the Summer of 2021, I took on the task of generating growth curve data for over 100 strains of anaerobic bacteria from the mouse gut microbiome. I had just moved back to the US from Sweden to start a new position as a Research Scientist in the Carini Lab at the University of Arizona. The prospect of collecting optical density (OD=600) data for dozens of gut microbes sans oxygen, was almost as challenging as the 112°F Sonoran Desert heat!

Share

I found myself trying to think of how to generate growth curves without cramming a huge standard-sized plate reader into our anaerobic chamber. Even though our COY chamber is a “double-wide” model and affectionately named Big Ben, it would still be a tight fit to put a normal-size plate reader inside! In my desperation to find a simpler solution, I sent out an SOS to the trusty Twitterverse asking if any of my fellow microbiologist tweeps had any ideas. In less than a day, Professor Rita Tamayo of the University of North Carolina, Chapel Hill, messaged me about her recent success with compact plate readers from Cerillo. It was a eureka moment of sorts, demonstrating the power of social media in fostering scientific collaboration and innovation. Within days we had ordered two of the Stratus microplate readers. Finding accompanying tiny plate shakers that fit in the standard COY incubator was a bit more challenging. The issue was the depth of the incubator was quite shallow, and most plate shakers would not fit. Eventually after some trial and error, we found the perfect match and we were off to the races. Please don’t let anyone tell you that Twitter (X) does not have its uses!

In the world of anaerobic microbiology, researchers often find themselves sweating away in enclosed glove boxes while painstakingly working to maintain sterility. This was certainly the case for two members of the team behind our recent study that developed a workflow for Cerillo's Stratus compact microplate readers.

Our findings were published in Microbiology Spectrum: “Cultivating efficiency: high-throughput growth analysis of anaerobic bacteria in compact microplate readers.” We demonstrated that the Stratus performs as well as standard microplate readers but with the added flexibility of operation inside the confines of an anaerobic chamber. “Working in the anaerobic chamber is tight, and we were constantly worried about contaminating our cultures,” recalls our tenacious undergraduate researcher, Makaela Valencia. “I also have short arms, so I had to get creative and use these super large forceps to reach some things in there!” she says, laughing. A huge fan placed at just the right angle also alleviated our rivers of face sweat and the steaming up of my glasses.

But what motivated us to pursue anaerobic microplate reader validation in the first place? Our goal was to simplify the lives of researchers, like ourselves, who study strict anaerobes. These compact plate readers, aside from being very cute, allow us to perform high throughput growth experiments in an anaerobic chamber. This avoids the hassle of collecting subsamples for OD readings, transporting them out of the chamber to be read in a standard plate reader. Beyond increasing convenience, our work has important implications for advancing our understanding of the anaerobic microbial world - from microbiomes to environmental systems. Being able to efficiently cultivate and monitor the growth of anaerobic isolates in parallel is a real boon for high throughput microbiology and opens avenues for new insights into things like gut microbiome dynamics, biogeochemical cycling of toxins like arsenic and mercury in anoxic ecosystems, syntrophic interactions of anaerobic microbes and much more.

As challenging as our hands-on research was at times, we validated our approach as a new tool for anaerobic cultivation workflows. We had to get pretty creative at points, for example finding ways to maximize our limited space by stacking reservoirs and tubes and using forceps fit for a giant, but in the end, we’re proud to have overcome the obstacles and to establish this capability for our lab and the field. From sweating it out in the chamber to transforming anaerobic cultivation studies, we hope that this behind-the-paper story showcases the work and perseverance that propels scientific advances. We are proud of our team’s published paper, but perhaps more importantly, behind its pages lies a journey of problem-solving, camaraderie, and fastidious dedication to life without oxygen; after all, as Ralphe S. Wolfe famously opined: Study of obligate anaerobes continually provides that extra challenge for all procedures, but for some of us, challenge is what it is all about (Wolfe 1999).

It's dizzying.

For years I made student loan payments without a strategy. I viewed student loans like a car loan or a mortgage: something that I needed to plug away at until it was gone. And the sooner I repaid it, the better.

Turns out, that wasn’t true.

As someone on the academic track, I am eligible for Public Student Loan Forgiveness (PSLF).

Congress created the Public Service Loan Forgiveness program in 2007 to encourage people to work in public service. If federal student loan borrowers make payments on their loans for 10 years while working in public service and meeting other requirements, Education forgives the remaining balance on their loans.

Work for universities falls under the umbrella of "Public Service". My years employed by universities as a postdoc and a professor counted. Now that I've made 10 years of payments, the loan balance is set for PSLF.

I had been in repayment for six years before I realized this. The PSLF game is very different than a mortgage or an auto loan. For PSLF, the goal is to pay as little as possible. This has the effect of “pushing” money forward toward forgiveness. To achieve this, there are strategies to help cut your monthly repayment cost. The strategy is not to pay as much as possible each month toward the loan.

Having a plan of attack is key for both PSLF and other loan repayment plans. Yet, the federal loan landscape is so dynamic that it's impossible to plan.

I found Student Loan Planner while searching for formulas used to calculate repayment amounts. At first, I subscribed to their free newsletter. The newsletter is very useful and became one I looked forward to reading each week. They are on top of the dynamic student loan landscape. They discuss what's happening in Washington. They explain how new laws will affect different repayment strategies. Below is a snippet of the most recent edition (March 2024).

After two years I booked a consult for a custom plan. I wish I had done it years earlier!

Here's how it works: You book a consultation and upload your student loan information. You meet with an advisor. SLP develops a plan specific to you. You follow the plan with periodic check-ins and follow-ups.

Subscribe now

A $300 consult saved my family tens of thousands of dollars in loan repayment savings. I've referred many friends and they have similar stories (see below).

It's never too early for early career professionals to begin thinking about the foundation of their financial future.

Share

There are inherent ethical and power dynamics in authorship decisions. Many unscrupulous players abuse this power to control who becomes an author, and how they are listed, and to sneak onto papers they contributed minimal effort toward.

We think that’s crappy! So we outlined Carini Lab authorship guidelines for transparency, consistency, and accountability. Please consider these free for public use.

Carini Lab Authorship Guidelines

1. We don't make detailed authorship decisions until the manuscript is almost complete.

2. We're liberal with what warrants authorship. Everyone is eligible for authorship, regardless of rank. Any preparation, provision, or action that directly contributes to the published work warrants authorship. We also include direct activities that made the published work possible, but may not be in the work. A key preliminary experiment for example. Contributions may include any of the following (or others not listed):

   1. experiments

   2. datasets

   3. analysis

   4. figures

   5. writing

   6. infrastructure

   7. expertise

   8. detailed discussion

   9. ideas

   10. troubleshooting

   11. organizing

   12. coding

3. All authors are responsible for the accuracy and integrity of their work. They should be able to defend their contributions.

4. The "lead author" takes primary ownership of the work as a whole. This includes making decisions on the figures, data analysis, writing, and revising. They conduct the bulk of the organization and writing of the manuscript, even if they did not do all the work. They submit the manuscript after the final review by Dr. Carini. They take part in revisions. They may or may not have conducted the experiments. They should understand all aspects of the presented work, even if others did the work. The lead author appears first on the manuscript.

5. In rare situations, "co-lead" authors are appropriate. Co-lead authors contribute equal energy to "lead" activities, as defined above. The authors and Dr. Carini will discuss the listed order of the co-lead authors. Co-lead authorship is an exception, not the rule and should be agreed upon by all parties.

6. We list contributing authors in order of their contributions to the manuscript. Authors listed earlier contributed "more" to the manuscript. Sometimes it is tricky to compare effort or authors have equal contributions. In these situations, we list authors in alphabetical order by last name.

7. The last author listed is the principal investigator. Sometimes the principal investigator is the lead author. In these cases, there is no special designatory role for the last authorship spot.

8. The principal investigator is the corresponding author on all works. Postdocs or senior researchers are always welcome as co-corresponding authors.

9. All authors will have an opportunity to revise the manuscript before submission. We limit this to an "open comment" period of ~2 weeks (unless other arrangements are made).

10. Authors that leave the lab before a manuscript is finished retain authorship.

11. You can opt out of authorship for any reason.

12. We expect lead authors to complete manuscripts in a reasonable timeframe after leaving the group. We'll define the timeframe in our exit interview. We'll reassign the lead author of abandoned works.

13. We do not look kindly on "sitting" on a manuscript. Manuscripts are the culmination of a tremendous amount of work, both in terms of person-hours and financial resources. Much of the funding comes from taxpayer dollars. Thus, it's imperative to prepare manuscripts as fast as carefully possible. If there is writing to be done, do it. We’ll consider the work abandoned if progress is not made in a timely fashion, outside of extenuating circumstances.

14. Acknowledgements are appropriate for people who have indirectly contributed to the research. For example, undergraduates often learn experiments through watching staff perform experiments. Watching does not earn co-authorship. Preparing media for an experiment does not earn co-authorship. Learning how to run a gel does not earn authorship. These are all more appropriate as acknowledgements. If there is a question about what earns authorship please ask your direct supervisor!

Does your group have extra guidelines? Did we miss something? We'd love to see them in the comments below.

Leave a comment

A Paper

Scott et al., published the minireview "Widespread dissolved inorganic carbon-modifying toolkits in genomes of autotrophic Bacteria and Archaea and how they are likely to bridge supply from the environment to demand by autotrophic pathways" in Applied & Environmental Microbiology.

This is a tour de force summary of how microbes interact with environmental dissolved inorganic carbon (DIC). DIC is sensitive to environmental conditions (notably pH). And there are a myriad of ways microbes interact with it biochemically.

The review summarizes the relevant microbial pathways and gene distribution across prokaryotes. They discuss the role of DIC related to engineering microbes for industrial or agricultural applications.

A Quote

Imagine taking a visit to a university, and having a chance to chat with everybody in an academic department to learn what they do for work. Office managers pull together many pieces to keep things running. Advisors advise. Lab prep people prep the labs. Research techs do research. Lecturers teach. And professors do, it seems, a little bit of everything? Some of us might do more research, some of us might do more teaching, but we all do a lot of the other stuff.

—Science for Everyone by Terry McGlynn

Science For Everyone

The meeting scheduling crisis is an indicator of a much bigger problem

Our calendars are so packed, there’s no time to address real priorities and generative work. This has been the theme of my past couple weeks, let me share four recent things that have brought me to this line of thought: a. I was on an advisory board meeting for an externally-funded faculty development project, which everybody thinks is a valuable endeavor. We got around an incomplete objective. What stood in the way? The faculty involved could not find time in their calendars to get together. It was simply an impossibility. Someone wrote a big grant, had big plans, and the thing didn’t happen because the calendar would not allow it…

Read more

2 years ago · 12 likes · 4 comments · Terry McGlynn

A Thought

Terry's post reminded me of an earlier version of myself in late 2019. I was about halfway into my tenure track position. And I was a mess. “Stressed” puts it lightly. This led to health problems (mental and physical). I was trying to optimize everything at work and home. Adding more blocks to the calendar in an attempt to be ever more efficient with my time. Adding activities to de-stress. Like training for an erg marathon (I did it, but in retrospect this was silly). Meditation. Adding more and more to the plate to find the right combination to unlock a more manageable balance.

Of course, this didn't help—it made things worse! I arrived at the same conclusion that Terry described in his post. I needed to do less, not more.

But how to do that when your job requires so much? I read It Doesn't Have to Be Crazy at Work. (And shortly after Rework—also helpful, but for other reasons). These books resonated with me. All this busyness doesn't have to be that way.

Leave a comment

I made deliberate changes to how I organized my day. Being spread too thin was gone in favor of large blocks of uninterrupted time. I completely changed how I ran the lab. And re-envisioned my purpose as a professor.

My mantra? Be good enough.

Academia tends to select a certain type of individual who can juggle all the balls. I learned that juggling act from years of feeling like no matter what I had done, it wasn't enough. I needed to do more. More papers. Better teaching reviews. More grants submitted. Happy lab group. Happy family. More, more, and more. So I became more efficient and accomplished more—at the expense of myself.

Academia, like other professions, can be a mountain with no summit. You think the pinnacle is around the corner, but as soon as you get there, there's more to climb. Because there is no top. There's always more to do.

Good enough is plenty. Good enough means acknowledging and accepting imperfection. That trying to juggle everything means that I’m focused on juggling, not the individual tasks themselves. I say "no" to more than I ever have. I focus on my physical and mental well-being way more than I did in 2019. If I'm not 100%, I can't be the best dad or husband. If I'm not 100%, my mind isn't in the game at work and my lab and my students suffer. This approach leaves me with less time to spend on work-related activities. Which means I can't do as much. I've learned to be OK with that. Good enough is more than enough.

Stay in the loop, subscribe!

Humans observe the world around them. As scientists, we investigate the current state of knowledge about these observations. We develop ideas for what might explain an observation. And we design an experiment to test our idea, carry it out, and analyze the results. Finally, we assess the results in the context of the original idea and the cycle repeats. Over time, these explanations become facts (more likely to be true, but still a probability).

Stay in the (inoculation) loop

Making the jump from an idea to an experiment is one of the hardest parts of science. Ideas are conceptual. Experimental data—numbers, charts, figures—are concrete. How are conceptual ideas translated to concrete data? How are promising ideas distinguished from dead ends? What details are important? Which details are not? Which methods should one use?

Experimental design is a critical junction for all research projects. It is the bridge between abstract ideas and the hard-fought reality of the work. Good ideas are abundant. The availability of time and money to conduct experiments limits scientific insight.

How do we choose which ideas to pursue? How are experiments born from ideas? We developed an experimental design framework inspired by Ryan Singer's Shape-up philosophy. Framing experiments in this manner helps us identify which ideas to chase, and which to set aside. The framework relies on three levels of design with varying granularity: coarse, intermediate, and fine.

The coarse scale contours the idea from the original observations. Ideas are complex multidimensional monsters. Ironing out this complexity is critical to producing interpretable results. Our first step is to simplify these complex concepts into something tangible. We call this process Shaping.

Shaping is a process that combines our laboratory expertise with the tools and resources we have available while taking into account any limitations of time or money. It involves experimenting with different designs to turn our ideas into reality. The process is highly creative and allows us to think big to bring our ideas to life.

When Shaping a project, I sketch the project out. Because the ideas are complex and rough, and I initially want to keep them at a coarse level, I'll use a writing utensil with a wide tip. This imposes a physical limit on the amount of detail I can include in the sketch. There are wide-tipped whiteboard and paper markers. I like the flexibility of the pencil and have used a lead holder with 5.6 mm lead. Recently, I've migrated to cheap carpenter's pencils. I’m sure crayons would work well too. If you're more tech-savvy, you might use an iPad notetaking app with a wide line. These implements allow for the right amount of detail while Shaping at the coarse level.

Shaping involves mapping the experimental design, time scales, controls, and high-level data analysis. I sketch workflows. Temporal dynamics. I'll sketch examples of what the end figures might look like. I'll highlight parts that might be tricky to pull off.

The goal of Shaping is to uncover the essence of the idea. If I run out of room on the paper or the whiteboard, the scale is not coarse enough. I'll look for things to take away while getting at the core of the idea. Shaping should feel a bit unrefined when it’s finished.

An idea that has taken shape provides researchers with ample opportunity to make their mark on the project. They can rely on their experience to fill in the intermediate and fine-scale details. The complex idea is now well-defined, with clear boundaries in place. We have identified the crucial aspects of the idea and either noted or avoided them altogether. The core of the idea flows smoothly. When experimenting, new ideas and interesting avenues inevitably emerge. By laying out the broader vision ahead of time, we confine the experiment to what we’ve mapped out and limit it accordingly. This approach lets our researchers know what is in play and what they should avoid doing.

The intermediate scale adds resolution to the coarse scale. This includes important details like the number of replicates to obtain. The sources of organisms or samples. The statistical tests to use. The order of operations and timeline estimates. Who to involve and when. The essential experimental controls, and those that are nice to have. Often, I’ll get a finer pencil in a different color and overlay these details onto the coarse-scale sketch. This is when the real scope of the experiment begins to take shape.

We’ll often design a pilot experiment from these intermediate-scale experimental plans. Pilot experiments are useful if we're venturing into a new area of research. These small experiments teach us how to do the experiment before setting it up on a larger scale.

The fine-scale is the mortar holding everything together. It fills the remaining gaps and finalizes important decisions. How much substrate to add. Which protocol to use. Materials needed. Statistical packages or code to use for data analysis. The figures to generate. We rarely think of these details ahead of conducting a pilot experiment. Why? Unexpected issues always arise, and you can never cover all the bases ahead of time. Data are noisy; we need more replicates. Cells grow slower than predicted; we need to adjust the timeline. The plots conceal a data distribution you hadn't anticipated; we need different plots.

At this fine-scale resolution, decision fatigue can (and often does) set in. There are so many decisions to make, and they are often interrelated. We all want to nail it on the first try. But that rarely happens. Every decision made results in a splay of downstream what-ifs. Spot this in your thinking with thoughts like “But what if…?” “How can we know that…” “I worry about…”. The fine details are often the most overwhelming part of the experimental process. But remember, you can’t cover all bases. Ever.

Once we have all three levels filled in we’ll assess its feasibility. Often, pilot data will help us to decide what to pursue and what to shelf. Budgetary or time constraints factor in as well. We continually ask “What experimental attributes can we take away?” before the whole thing falls apart. This simplifies everything and is the antithesis of the obsession with new methods. Once we have a shaped idea with all the details filled in that’s within our scope…it’s game on!

Do you have a structured way of thinking through new ideas? How do you develop experiments? What tools or tricks do you use? Please share in the comments!

Wrong.

Before diving into the details of what researchers cost, there are some concepts that are critical to understanding how people are funded. Different institutions might have different language around this, but the core ideas are likely similar.

Full Time Equivalent (FTE): FTE is used to explain an employee’s percent effort on a funding source relative to a 40-hour workweek. For example, a full-time employee working 40 hours would be at 1.00 FTE (that is, 100% of a full-time 40-hour workweek). In contrast, a graduate student at 0.5 FTE equates to 20 hours of work (50% of a full-time 40-hour workweek).

Employee Related Expenses (ERE): This is the cost of an employee beyond their base salary, usually expressed as a fixed percent of their base salary. This is also known as fringe expense or fringe rate. Wonder why you get paid time off (sick time and vacation)? Or where your retirement or health benefits come from? They come from ERE. Importantly, the ERE percentage varies by position classification, reflecting the differences in the benefits associated with each classification (more below). These costs are typically paid from the same source that funds the employee’s salary (grants, start up funds, departments, etc).

Understanding ERE and FTE is important when constructing proposal budgets to fund the researchers conducting the research. It might seem like the order of operations in constructing a grant proposal is to determine what cool science you want to do first, identify what types of researchers are needed to do the research, then construct a budget around the science. While that’s partially true (you’ve got to have an idea to begin with), it’s really a waltz between the budget, the science, and the duration of funding. Often, you’ll have to adjust the scope of the project—or the types of personnel funded by the proposal—to fit funding limits or to make the project doable in the timeframe of funding offered.

Below, is a breakdown of what different types of personnel cost as they currently apply in my current department at the UArizona. While some of the expenses below are set at the university level (ERE, for example), others (like graduate student salaries) vary at the department or college level. These values (percentages, salaries, and other policies) often vary—sometimes greatly—across universities.

1. Principal Investigators (PIs; ERE rate 31.9%): Many professors that run research labs are on 9-month contracts. That is, they are paid for the 9-month academic year. They can obtain support for the three summer months as “supplemental compensation” from research grants at the same salary rate. For example, a professor that makes $80,000 on a 9-month contract, can bring in (up to) an additional $26,666 on grants should they be fortunate enough to have ample funding. The base $80,000 is paid by the University, the PI must solicit the remaining funds from grants.

   In addition to the PI’s base supplemental compensation salary, the budget includes the ERE for their salary. The UArizona ERE rate for PIs is 31.9%. So, one month of an $80,000 salary is ~$8,889 ($80,000 salary ÷ 9 months). ERE is an additional $2,836 ($8,889 × 0.319 = $2,836). For a total of $11,725 per month ($8,889 + $2,836 = $11,725). So, it costs nearly $12,000 per month of supplemental compensation. If the PI’s base salary is higher, this cost is higher.

   Many funders have upper limits on how much time (in terms of months), or dollars, can be spent on supplemental compensation for the PI. For example, the National Science Foundation limits this to two months of support across all NSF grants. Additionally, paying yourself above your rate is commonly not allowed. For example, in the above $80,000 example, you could not pay yourself more than $8,889 base salary per month.

2. Staff positions (ERE rate 31.9%). It’s often said that good technicians are hard to keep funded. This is because they are in the highest ERE class and a good tech should make a good salary. A double whammy. Using similar math as above, a technician that takes home $65,000 costs the grant $65,000 × 1.319 = $85,735. This is often why technical staff salaries are often pieced together across several grants. For example, a well-funded principal investigator may have three awards and split 33% of the technician’s effort (i.e. 0.33 FTE) across these three grants, such that each contributes equally to fund the researcher. I’ve heard that the ERE rate for similar positions at some universities can be as much as 80%, which makes it difficult to fund technical staff long term.

3. Postdocs (ERE rate 17.6%). Postdocs at UArizona do not pull the same ERE rate as technical staff because they are offered reduced UArizona benefits compared to other appointed professionals. (Something that I do not support). A postdoc making $65,000 take home pay costs a grant $65,000 × 1.18 = $76,700. That’s $9,035 less per year than a technical staff with the same take home pay.

4. Graduate trainees (ERE rate 13.0%). Graduate students are considered ‘trainees’ at most higher education institutions. As such, their benefits are typically greatly reduced compared to staff and postdoctoral researchers. This translates to a substantially lower ERE. Additionally, many institutions place bounds on allowable graduate student salaries and FTE. During the academic year, UArizona graduate students are capped at 0.5-0.66 FTE, depending on their citizenship. During breaks, the limit increases to 0.875 FTE, although individual PIs may choose to pay students at a lower rate. In my department, a doctoral graduate assistant is paid from a 9-month 1.0 FTE base salary of $45,964. But during the academic year, they are capped at 0.5 FTE, so their take home salary for the academic year is $45,964 × 0.5 = $22,982. If they work in the summer at 0.875 FTE (406 hours), that is an additional $13,486, for a total annual take-home salary of $22,982 + $13,486 = $36,468. Then we add 13.0% ERE on top of this: $36,468 × 1.13 = $41,208. The cost of graduate students doesn’t end here, as we also pay their tuition on top of their base salary and ERE. At UArizona, the graduate student tuition for is typically paid from a grant. The current annual tuition rate for ≥0.5 FTE students is $12,718. For every year of salary ($41,208), we need to add $12,718. That totals to $53,926 per student, per year. Tuition is paid even after their formal coursework is finished and the student is taking research or thesis credits. Some universities waive tuition for graduate students, which makes funding their training a more economical.

5. Undergrads (ERE 2.0%). Undergraduates are typically hourly employees and capped at an upper FTE limit. At UArizona these limits are 0.5 FTE during the academic year and 0.875 during the summer. Their fringe rate is low because their positions as student workers are not associated with fringe benefits. We start undergrads at minimum wage and increase their wages as they grow in their skill sets, usually on an annual basis.

Subscribe now

Below is an example of the personnel costs for a proposal from a few years ago. The total proposal value was ~$675,000 and the personnel cost alone was just under $410,000. That is, funding the person-power was 60% of the total budget.

As we climb through our academic careers, our perception of the cost of science changes. As a graduate student, I envisioned a $100 reagent as a crippling cost for the lab. Looking back, I probably thought this because $100 would have been a massive cost to me in my personal life at the time. In reality, that $100 is minuscule, relative to the cost of the person doing the research with that reagent.

An example of this thought process at work can be seen looking at the cost of high-throughput DNA extractions in microtiter plates vs classic DNA extractions in single tubes. Single-tube DNA extraction kits are ~$0.21 per reaction cheaper than the microtiter plate-based extractions. But when you factor in the amount of time saved across 96 samples viewed through the lens of the cost of someone’s time spent doing the extractions, the cost per reaction is much reduced—even if undergraduates are conducting the extractions—simply because it takes less time to run the extractions in the high-throughput format. (Of course, this example is only a meaningful savings if you have many DNA extractions to conduct).

As we climb through the academic ranks, we’re taught (both directly and indirectly) that our time is not valuable. But it is. I always ask “how much time will it save you?” when someone asks for a new tool, no matter the cost. And I try to weigh the utility of that tool over time—in the context of personnel hours it saves—when deciding whether to make the purchase.

So often we forget to factor in the most expensive cost of running a lab—people— when making decisions about what tools to buy or not.

The structure is inspired by how 37 Signals structures their work into 6-week cycles (explained in Shape Up). While we are not software developers—and few projects fully fit within 6-weeks (for biological reasons)—this approach provides a clear structure that works within the context of our academic year, that is effectively 4-5 6-week cycles long. It’s also very manageable for students to think in 6-week increments. Longer than that is too abstract. Shorter than that can be too granular.

What follows is a nearly-complete version of our rotation introduction that we’ll send to the students at the beginning of their rotation. It describes both the structure of the rotation and our evaluations.

There are likely dozens of ways to host students in lab rotations. We’ve not tried this before and there will likely be unforeseen pitfalls. We’ll adjust on the fly as needed.

Today is SAR11’s scientific birthday—the first paper describing the bacterial 16S rRNA clone from SARgasso Sea clone number 11 (SAR11) was published 33 years ago today.

Most of our current knowledge about the physiology and ecology of SAR11 is summarized in a recent review paper. We now know that these marine heterotrophic bacteria numerically dominate marine waters from the surface to the deep ocean globally. And they are the most numerically abundant cells on the planet.

SAR11 are masters of minimalism, coding for only ~1,300 genes, and their genomes are still among the smallest for a free-living organism. They lack abundant transcriptional regulators and as such minimally regulate transcription of their genome.

Share

Many of the foundational discoveries on SAR11 were initiated in the 1990’s well before the deluge of YouTube videos. However, we are extremely lucky that in the late 1990’s PBS published a series of videos titled “Intimate Strangers: Unseen Life on Earth.” In one of these episodes—Oceans of Microbes—they described what is effectively the discovery of SAR11. The video can be seen here (at the 25:23 time stamp). This is a rare contemporaneous snapshot of a foundational discovery in microbiology. [Note: this video used to be much easier to find. It appears that the link above is the only instance…and who knows how long it will be posted]

More recently (2019), Dr. Giovannoni was a guest on the ASM “Meet the Microbiologist” podcast discussing SAR11 and SAR202 with Dr. Julie Wolf.

As a doctoral student, I stumbled upon a defined artificial medium to culture SAR11. SAR11 had been cultured, but was dependent on natural seawater as a base for growth. That meant every few months researchers would need to hop on a ship with 30 20L carboys and head off the Oregon coast to collect seawater. Occasionally, we’d have Seawater shipped to Oregon from the Sargasso Sea in the Atlantic. Identifying the best “batches” of seawater for SAR11 cultivation was somewhat of a competition. Every student or postdoc would cordon off a set of carboys so they could conduct reproducible experiments without variability introduced by the seawater batch. That is, certain ‘vintages’ of seawater performed better than others.

At the time, I was trying to identify “the missing nutrient” that would allow us to culture SAR11 on synthetic medium. The prevailing theory was that although a handful of essential requirements had been identified—reduced sulfur, glycine, and some low molecular weight organic acids— there was something still “missing” from their growth medium that was preventing us from cracking the 1.0 × 10^7 cells per ml threshold. I thought it might have to do with nucleotide metabolism. Specifically nucleotide monophosphates. So I set up a bunch of experiments testing these ideas.

I set up three experiments: one with 100% artificial seawater (ASW), one with 50% natural Sargasso seawater + 50% ASW, and one with 50% natural Oregon coastal seawater + 50% ASW, the full complement of known requirements, and assorted combinations of nucleotide monophosphates. I started the experiment on 12-Nov. 2008: below is a snapshot from my lab notebook.

I called this experiment the “3 Waters” experiment because I used 3 waters (ASW, Sargasso water and Oregon water). I measured the cell densities of the flasks for a bit over a month. While I can’t find the growth curve from these experiments, I do recall that the cell densities did not change at all over those 4-5 weeks. But cells were not dying and that stood out to me as important. Then I gave up. See my 18-Dec 2008 notebook entry:

I placed the flasks back in the incubator and forgot about them until I needed some incubator space. I remember gathering the flasks and was right at the sink, about to dump them down the drain. When I thought “I’ll check these one more time”.

In the intervening time between my “final counts” and me getting ready to dump them, SAR11 had started growing in the 50% seawater 50% ASW flasks. These cultures were passed to new 100% ASW flasks and they grew. We verified it was SAR11 via RFLP:

The rest is history.

Over the years, dozens of scientists around the world have contributed to the now vast knowledge base on SAR11 microbes—too many to list here. While I haven’t kept up with all of the ongoing SAR11 research lately, I feel very fortunate to have contributed a small slice to the ever expanding SAR11 knowledge base.

Happy Birthday SAR11!

We have several projects related to marine microbes that might be a fit for a postdoc, including:

1. Energy limitation in marine Thalassospira and/or Roesovarius. This project would include applying chemostat-based cultivation, genomics, and other -’omic approaches to robustly growing marine microbes.

2. Isolation of microbes from marine sediments. We are developing methods to culture a broader array of microbes from soils and sediment communities relative to traditional methods. This would be an excellent project for postdocs with interest or experience in genomics, and metagenomics (including MAG assembly) looking to learn microbial cultivation skills.

3. Characterization and further inquiry into a variety of coastal sediment enrichments. We set up a series of experiments that have produced an array of enrichments of assorted bacterial and archaeal taxa. Some enrichments contain high relative abundances of novel microbial taxa.

4. Your independent ideas. We are interested in bringing in postdocs who both bring fresh perspectives to our group and are interested in learning new skills from us.

   

The Carini Lab can sponsor one application for this fellowship. If you are interested in us sponsoring your work, send the following by March 31 to Dr. Carini (paulcarini@arizona.edu):

1. A summary of your dissertation research. (1/2 page limit).

2. A summary of your longer-term career goals. (1/2 page limit)

3. Which project you are interested in (listed above, or your own project), and why you are interested in joining our group. (1/2 page limit)

4. A copy of your CV

We will review these applications and make a decision about which one to support shortly thereafter. Final applications are due to the Simons Foundation on May 7.

Subscribe now

There are two parts to science. The first is conveyed nicely by the scientific method. This is where hypotheses get tested in laboratories with fancy equipment like lasers and test tubes. The second is conveyed by text and figures, in documents designed to acquire research funds or communicate research results. No question—writing is central to the scientific process. Yet, writing is a roadblock for science trainees. The reasons for this are complex. Perhaps junior scientists are focused on acquiring new research skills. Perhaps they already “know how to write.” Or it’ll be easier than experiments. Maybe, they’d rather be doing anything else on Earth.

Whatever the cause, trainees face a writing mountain. Writing is hard work. We hope it’ll be easy. That it’ll “come to me.” But it’s hard. First you write. Then revise. Experiment with sentences. Start over. It’s messy. It’s moody.

We procrastinate because it’s hard. With a lack of a deadline, this goes on forever. With a deadline, we binge write—usually getting serious at the last possible minute so the piece will be finished in time. But just barely. Neither scenario is ideal.

There is this popular culture around writing that imposes a sense of guilt if you’re not doing it. “You should be writing.”  This is a path to anxiety, feelings of shame, insecurity, and unworthiness. Not excellent prose.

Like laboratory procedures, writing is learned by doing. If you are not writing, you are not learning how to write. Learning how to write is a lifelong process. Just like your first experiments in the lab were messy, so too with writing. Your skill will improve with time.

Over the years, I’ve shared my own tricks (below) with trainees and colleagues. This is a system that I designed to promote writing success on my worst days. (On my best days, I don’t need help writing).

1. Writing is hard work. You can do hard things. Once writing is appreciated for the hard work that it is, your relationship to it will change. Instead of loathing it, you’ll learn to approach it systematically like any other problem. You might even learn to love it. (But it’s OK if you don’t).

2. Writing is a laborious form of communication. Most of us don’t need time to think before words come from our mouths during conversation (arguably some of us should). Writing is different. We need to play around with how the words work together. What preceded the current sentence? What comes after? Does it flow? Does it have rhythm? All of this makes writing hard (see point 1).

3. Push through the messy phase. There is a messy phase to writing. It usually occurs early on—but not immediately after starting. It’s the period when trying to identify the unifying thread that captures the essence of what you’re trying to write. This is sometimes conceptually presented as “your first draft will suck.” But it’s not really about the draft, per se. It’s more about the creative process of writing. There’s this idea that you should know what to write before you even sit down. This is nonsense! What you write is the result of an interaction between your mind, the page, and the data. Write. Read. Revise. Repeat. Let it evolve.

4. Schedule it. Stick to it. Do not double book writing time. This time is non-negotiable. If you’re new to writing, start by blocking off 20 or 30 minutes daily for two weeks. After two weeks, increase to one hour. Over the next several weeks try to work up to two hours daily. This is my maximum time limit, and I (learned to) like writing. A consistent two hours of writing per day is probably all you’ll ever need. I promise. Schedule this first and fit your remaining commitments around it.

5. Prioritize it. Don’t stick your writing time at times of the day when you’re low-energy (like after research time or meetings)—when you’re tired or hungry or bored. I schedule my writing time when I’m at or near my peak. I used this Chronotype quiz to help me find my appropriate time.  

6. Block distractions. Writing requires entry into a deep work phase. It’s impossible to do that with phones or computers pinging you. There are numerous apps designed to block access to distracting websites on your laptop. I block all social media, email, (un)productivity software (Slack, etc), YouTube, reddit, news sites, sports sites, etc. with the SelfControl app. Put your phone away. Seriously—silence it, turn the vibration off, and physically hide it.

7. Scope appropriately.  A paper can’t be written in two hours. A paragraph can be. Instead of outlining an entire paper, spend a few minutes defining what to accomplish in the current writing block. That’s it. There’s no need to reach further into the future than the current time block. The paper emerges as blocks are stacked over time.

8. Track it. In the beginning, track how many words you write. Below is a snippet of the spreadsheet I use to track my writing. If you’re data savvy, you can plot this data to determine which day-time combinations are your most productive.

   

9. What’s in bounds? I use my writing time to analyze data, generate figures, write, and edit research proposals and manuscripts only. Yes, there are a lot of other things to write. But my writing time is strictly limited to these items. For example, I schedule a separate recurring time block for the Un-Cultured newsletter.  

10. Find your voice. Often scientific writers try to sound a certain way on paper—the way they “should” sound. Don’t try to sound how you should sound. You should sound like you. How does your unique identifiable writing voice sound?

11. Read more. Pay attention to how the author writes. Analyze their paragraph structure. Contemplate what the author didn’t write. Why did they choose to write a sentence the way they wrote it and not another way?

12. Read about writing. There are numerous excellent books about writing (in broad terms), scientific writing, and habit building (which is integral to writing). I’ve listed several here.

There are infinite tips and tricks about writing. For me, it’s all about formulating the habit and sticking to it. Conceptually, these tips can be applied to just about anything, including coding, experiments, and other activities you might be less-than-motivated to carry out. I hope these pointers help get you motivated to write consistently in the future!

Share