Over the years, the lab and I have come up with various materials for science process (writing, presenting, reviewing, etc.). Here they are, loosely organized in a Google Doc. Hope it is useful!
Bit of advice for junior scientists: thinking about your experimental design is good, and most of us should do more of it, but sometimes just doing the experiment helps move things forward in ways that you can't anticipate, no matter how much reasoning goes into it.
Note to lab members everywhere: please please cc your PI on all emails to collaborators, no matter how mundane. It’s critical to avoid issues down the line.
Reviewed a paper and signed off on the revision; now seeing re-reviews from other authors that are as long as the original list of concerns. I'm not sure when multiple rounds of review became normalized behavior, but it really needs to get unnormalized. IT'S JUST A PAPER.
It's been great meeting with postdocs about the faculty position application process! Put together some collected thoughts on the process here. Main thing I've noticed: work on describing your research and *why it's important*. Concrete examples within.
I recently found out that editors will sometimes *purposefully* ask people from your "do not review" list to review. Dear editors, if you do this, please stop. There's a lot of very serious reasons that people exclude reviewers, and so it's critical to honor these requests.
Sorely tempted to just swear off publication entirely and let
@biorxivpreprint
be the official record of our work… would have so much more time for actual science…
Writing is an absolutely critical skill… and one that can be improved dramatically. Doing so requires not just repetitions, but deliberate efforts to improve. Some tips here:
What the German word for "I want to include a cool new result in a paper we are writing but am worried that it opens up too much new stuff for reviewers to ask about so we'll probably keep it out for a later paper that will likely never actually materialize"?
Been putting together some documents for writing. Came up with this list of "Words to avoid when writing" to be a concrete list of things to search for and replace, along with examples of why they're bad and how to correct them.
Starting to think that "but what about in vivo?" is a particularly pernicious form of gatekeeping: only rich and well-connected labs can afford to handle these sorts of editorial/reviewer requests, most of which really don't really advance science in a meaningful way.
I think it's very telling that this visual description of biology is organized into layers of "omes", which is essentially just what we can measure, not necessarily what is biologically important or meaningful. True models probably lie somewhere in between or across these layers.
In retrospect, it was naïve to think that we could understand the uniqueness & medical essence of human beings just by knowing their genome sequence.
There's many more important layers—multi-modal—information to help do that
@CellCellPress
open-access 3/5
I am struck by the apparent lack of correlation between lab organization and lab productivity/impact. I wonder if that's because, in our line of work, creative destruction provides more value than process optimization.
Wow, I still can't believe it! It is such an honor to receive this award. A huge thank you to all the members of my lab through the years—this award is for all of us! Also thanks to
@pennbioeng
for taking a chance on me those many years ago, and to
@PennEpiInst
and
@PennGenetics
!
Am I the only one who measures lab spending in units of dollars spent on failed antibodies?
Like: "Oh, don't worry, that's just the cost of a failed antibodies, could have be worse."
Scientists often resort to stock critiques (“But what about in vivo” “drugs are non-specific” “RNA and protein don’t correlate”). Think harder. Real alternative hypotheses require as much care and attention as the primary hypothesis in order for them to really contribute.
Not sure who is giving this guidance, but PhD students: we don't need to meet to discuss whether or not I should be on your thesis committee. Just send a description of what you do and why I might be helpful. I can decide based on that.
Faculty application season is starting up! Here's some thoughts on the process:
Part of Tools for Science:
Also, my offer stands for 1-1 mentoring to anyone who wants to chat, just send me an email and we'll find a time!
Dear editors: I get that journals don't want to commit to an "acceptance" until the bitter end. However, please consider trainees for whom putting "accepted" on their CV can really make a difference. It's gotten to the point where I can't even tell when it's "accepted" myself.
PSA: We’ve been buying DI water from Target for a while now instead of paying for a Millipore service contract. No problems so far, and saving considerable money.
Dang… Dear ChatGPT: "Point out ambiguities in the following scientific protocol: [our smFISH protocol]"
Have been asked many of these specific questions…
Can I just say that after having a (really solid) paper working its way through review for *3+ years* and *still* dealing with baseless criticism from reviewers, eLife sounds very attractive to me.
This is so crazy. Take out the queen, and workers ant fight, with the winner *transforming* into a queen, gaining additional lifespan. Using scRNA-seq, authors show these "queens" have more glia, suggesting that that helps increase brain health to enable longer lifespan.
Our single-cell ant brain paper is up on
@biorxivpreprint
. It's the first single-cell study in a social insect brain, and we're really excited about it! (1/3)
Pro tip for PhD students: Don't stress over experiments/analyses your thesis committee suggest. Don't ignore, but it's not a requirement. You and your advisor are the experts. If it helps, keep in mind that most committee members forget their suggestions in about 10-20 minutes.
The most blatantly self-serving article I've seen in recent memory. The reason for loss of public trust in science is… that we don't treat editors with enough respect? Yes, editors should be respected. No, that has nothing to do with loss of trust. 1/2
Word on the street is that chromatin accessibility is better than RNA-seq for discriminating cell state. Any favorite refs for showing this result rigorously? Or is it just a "PCA looks more spread out" kinda thing?
Remind me again why I do this? (Done ~100 in the past 7-8 months with people at all stages from around the world.) Oh yeah, because it turns out that I actually really like helping people :). Anyway, if anyone is still left to take me up on it, the offer still stands…
Re-posting: happy to chat about career advice/science/whatever! Email me to find a time. Particularly happy to chat with those from diverse and non-traditional backgrounds!
Another bit of awesome author-centric goodness from
@CellSystemsCP
. Showing that journals can either wring hands endlessly about the needs of authors, or just listen to them and do something about it. Thanks, Cell Systems, for showing the way!
Every so often, the question of whether to normalize RNA counts by cell volume comes up in lab. Default answer: "yes, normalize by volume". Why? It's not just that RNA correlates with volume. It's that the relationship is *causal*. If you increase volume, RNA goes up. Thread 1/
Kinda weird how when you write a paper about some results, you make it sound like they provide the definitive answer, but when you write a grant on those same data, they are just the tip of the iceberg.
Friendly reminder that studies aren't "observational" or "mechanistic". Mechanisms are models that are based on observations. Some observations may eliminate some number of possible mechanisms. That is all.
We need to take a hard look at some of our core processes in academia (publishing, grants, paperwork) and give them an overhaul, or else face a generation of talent saying—rightly—“to hell with this!”
The majority of senior academics still don’t understand how fast this is coming. This isn’t about trainees pursuing industry when academia doesn’t work out. Many of the best trainees are looking at academic life and deciding, quite sensibly, that there are better options.
Seriously—many scientists and trainees are getting a very warped view of the world and science from twitter. It’s an amazing resource in so many ways, but be very careful what voices you let into your head. My 2c.
Graduate student: what topics would you like to see covered in a PhD "skills" class? (Already on deck: how to give a talk, write a paper, write a grant.)
Okay, here's how you do the QR code for feedback:
1. Make a Google Form for your talk feedback. Here's mine:
What I do is make a pre-filled link (gear top right) where you pre-fill the particulars of that talk. Then…
Yes. The scope creep for faculty jobs has made this job impossible demanding. We are simultaneously asked to do more of essentially everything for everyone else, but at the same time set a good example for work-life balance. At some point, something has to give.
Reflecting on posts of post-holiday stress: I think academic PIs (especially juniors) simply have too many jobs to do for 1 person. We need to be administrators, scientific leaders, teachers, someone who fights for money and even psychologists at times. This is objectively hard.
Probably the best advice one could give to someone prepping a fellowship application is to give them a stack of fellowship applications and give them 36 hours to review and rank them all.
Anybody else feel like response to reviewers have gotten longer and longer? Exhausted at the prospect of another 40+ page rebuttal to the collected works of Reviewer 3.
Reposting my offer: 1-1 mentoring by phone on anything science/career related! Have talked to undergrads to tenured faculty and everything in between. Email me to find a time. As always, very happy to chat with people from diverse backgrounds.
Reposting my offer for 1-1 mentoring on, well, just about anything! Faculty and grad app season is upon us, but anything goes. Email me for a time. As always, very happy to chat with people from diverse backgrounds.
ChatGPT calls into question the very definition of plagiarism, which itself is founded upon (false) notions of agency and ego. After all, we are all just ChatGPTs with a different training set, right? Those words in your head came from somewhere, by definition not "yourself".
The usual prescription for burnout is rest. While that sometimes works, I've also found that sometimes finding a new approach to work is actually more helpful in breaking out of bad mental cycles. Renewed purpose and vigor sometimes help me more than rest.
Sad that I don't do enough real work to justify getting myself a new M1 Max/Pro laptop. Reminds me that my job now basically requires nothing more than an entry level laptop for form filling and light text editing. Wah.
It occurs to me that some labs are best for being the first to do something everyone else is doing, while some are best for being the first to do something nobody else is doing. Both are fine, but choose your lane wisely.
I feel as though the state of a cell can be pretty accurately discriminated from some other state of a cell by virtually any sufficiently high-dimensional profile. If you want to know what that state *does* solely from the profile, well, I think you're out of luck either way tbh.
Paper submission is becoming almost as bureaucratic as grant submission. Feel like labs will soon need an admin just to handle all the assorted forms and uploads.
Blog post: It seems there is surprisingly little guidance out there on how to reintegrate trainees into lab after a mental health leave. Here are some thoughts on the topic from four anonymous trainees in a guest post:
One thing I've noticed: chalk forces you to cover way less material. That's a good thing! PPT enables you to cover WAY more material than anyone can usefully absorb. PPT is the presentation that makes you feel smart, not your audience :).
The night before my talk at a conference last week, I decided to make it a chalk talk, instead of using my slides. Not being tied down to slides allowed me to jump around as questions came up & our discussion led to new insights. Why did we let PowerPoint take over our meetings?
People do postdocs for all kinds of reasons. It is all good. Gather information, make the choice you think is right, and know that you can change course if needed. Many life twists are harder to predict than we think, so we tend to undervalue enjoying the moment.
Someday grad students will realize that being a postdoc sucks unless you’re academia or bust.
The market will reward PhDs with great salaries and great science. There’s no reason to do a postdoc unless you want to be a professor.
Pro-tip: keep a list of people in your field. Make sure the list is diverse and filled with nice people. Keep handy for whenever you are asked to name people for something (talk, prize, whatever).
For YEARS we've had occasional issues where someone leaves the fridge/freezer door open.
@laurenbeck131
's brilliant solution? Some shims to raise the front of the unit, letting gravity do the work for us!
In my experience, RNA-seq (while better than qPCR) is still not particularly accurate. See here for comparison to RNA FISH. And that's on a *log* scale. Accuracy is actually quite poor, even if it gets broad strokes right.
New preprint from our group. Q: How accurate is bulk and single-cell RNA sequencing? A: Not as accurate as we expected. In fact, at the single-cell level, PCR bias has a significant effect. We quantify this experimentally. A 🧵
#SingleCell
NIH grant amounts in real terms are reaching low enough levels where it can be hard to generate sufficient data for getting the next grant. This risks hollowing out the NIH middle class as those who can attract private funding will be able to put together stronger NIH grants.
Interview question: "What is an NYtimes headline for your work?"
Overly honest answer: "Local man turns grant money into sequencing data into a PDF into more grant money.”
Everyone always says to write the results section of paper first. I think it’s better to start with introduction. Otherwise, it’s like giving an answer without knowing the question.
The arguments presented here against reuse of language in methods section are unconvincing. Yes, it is certainly possible to rewrite a protocol over and over again. Is that a valuable exercise in scientific writing ala Flaubert? No way. We're not writing Madame Bovary here.
Seeing a whole lot of folks the last couple of days arguing that "self-plagiarism" in Methods should be allowed and we should do it even if it isn't and "there are only so many ways" to describe a Method.
So I guess it's time for this again:
In giving advice to junior faculty, I’m thinking it’s best to replace “Doing [xyz] was the key to my success” with “I did [xyz] and at least somehow didn’t fail”.
Wow, truly spectacular talk by Lea Goentoro today in the
@pennbioeng
seminar! Showed limb regeneration can be induced across multiple species through nutrient and insulin supplementation. Preprint here, and check out this abstract!
So happy to announce that
@Caroline_Bartma
is the winner of the Saul Winegrad Award for Outstanding Dissertation for the Penn Immunology Grad Group! Proud to have worked her with
@blobelgroup
. Jokes about puns and figure alignment aside, a great recognition for a great scientist!
Major lab fire on our floor last week. Overheard faculty talking about making sure that mental health services are available to those who accidentally started the fire so that they don't feel stigmatized and traumatized. A good reminder that academia can be a pretty nice place.