Dr. Jesse Coker

@HangZheng855161 @AChakrabortyPhD @BoudkerLab Agree on all fronts! Especially since many SLCs (by evolutionary design) move charged molecules, the active site inhibitors frequently have terrible drug-like properties. These allosteric sites can give you hydrophobicity and selectively.

@jcoker10 @AChakrabortyPhD @BoudkerLab Mechanism work on old SLC inhibitors keeps quietly resetting transporter selectivity. Allosteric handles look like the more rational route for SLCs the way they were for GPCRs.

Alongside @AChakrabortyPhD and @BoudkerLab, we disentangled mysterious allosteric inhibition mechanism of an old SLC1A1-selective inhibitor. We made some new analogs with better anti-ccRCC activity. Like GPRCs, think outside the orthosteric site for SLCs! link.springer.com/article/10.103…

LeBron was drafted when I was 8 years old. His 2007 Finals run was 6th grade. First championship in 2012 was junior year of highschool. 2016 3-1 finals comeback, junior year of college. 2020 bubble chip was last year of grad school. I now have a CHILD. AND HES STILL DOING THIS.

@jamessinka @Oliver__Hahn @claudeai I use it to check my own hypotheses or help me assign a prior probability to two mutually exclusive hypotheses. If you guide it to the right data it’s excellent at stress testing. I also use my own datasets where it’s helpful which is often best!

That's helpful! Are you checking your own hypotheses or finding gems across data? In your experience, how good has Claude been for either (checking or novel production?) Do you use it anywhere else besides public data intake? It still has room for improvement in materials science for me.

For a hackathon project in March, I assembled about 90,000 T cell receptor (TCR) sequences from 8 independent sources in about 1 day. TCRs are a little more exotic than classic RNA-seq data and thus stored in all sorts of different formats and databases (i.e. a pain in the neck). The agents handled retrieval from VDJdb, McPAS, TCRAFT, TRAIT, plus raw sequencing data from published papers re-processed from scratch. Every source used different schemas, different column names, different ways of encoding the same receptor. My job was knowing which datasets mattered and what "clean" means for TCR data. 3/🧵

My heart-felt congratulation to this year’s winner of the #AACR26 Award for Outstanding Achievement in Chemistry in Cancer Research Cheryl Arrowsmith. Well deserved for a pioneer of open science, high quality chemical probes & high throughput structural biology. @thesgconline

@jamessinka @Oliver__Hahn @claudeai Wish I had something concrete. Workflow is (1) read literature and aggressively save interesting supplemental tables (2) scour public repositories like GEO, ARCHS4, DepMap, Pharos, ect for large-scale data (3) let Claude cook for 👍/👎 verdict on new hypothesis.

@jcoker10 @Oliver__Hahn @claudeai That's awesome. Care to share more about your workflow so other scientists can do the same?

@Oliver__Hahn @claudeai Yes! Agree completely. The great thing about digging into public data is that I can also no-go ideas very fast and save the wet lab work too.

Totally! I get frequently surprised with folks wanting to deliver me the supposedly complete, top-to-bottom hypothesis AI machine, while THIS is actually driving productivity up across research. And the semi-interactive nature feels (to me) a lot more how I actually do science: not by a single, mega whiteboard session that delivers experimental plans for 3 years but through much smaller but frequent planning phases as results come in.

@Oliver__Hahn This has been my primary use case with @claudeai Claude Code for the past three months and it is transformative. Converts every thought from “boy I wish I could do this analysis” to “this will be done in an afternoon”. It’s a clear, defined, unsexy way for LLMs to advance science

One thing worth saying: most of this data was generated with public funding. It should be accessible. Often it isn't. It's buried in supplementary tables, poorly annotated, deposited in formats nobody maintains. These tools don't fix that incentive problem, but they dramatically lower the barrier to actually using what's already out there. And it matters more than ever to properly cite and credit the datasets you use. The easier it gets to access someone's data, the more important it is to acknowledge their work. If you're building AI science tools: this is not where you save on tokens - include those datasets & meta-analyzes in your scientific reasoning workflows.🪙🪙🪙

@KJangkeun Drug discovery focused structural biologist here—very cool. I have a growing history of publishing negative data, it’s a real goal of mine despite the annoyances (most recently doi.org/10.1021/acsche…). Would love to connect and discuss how to grow this great tool!

5. Working in drug discovery, clinical trials, structural biology, or AI evaluation? Would love to connect. #AIforScience

Excited to share that NegBioDB has been accepted into @AnthropicAI 's AI for Science program! A database of biological negative results: drug discovery failures, trial terminations, and PPI data. About 63M records spanning multiple biomedical domains.

I tweeted this at a weird time but now it’s the evening so you are required to gaze upon my suffering 😅

We tried to make merged fragments at the NSP14 ExoN active site—now you don’t have to! S/O to @XChem_DLS and @DiamondLightSou for sharing hits! We strive to publish ALL our data, especially negative. Fantastic that @AmerChemSociety supports this. pubs.acs.org/doi/10.1021/ac…

What's the role of MCL1 in chemoresistance? A new preLight from @kparashar1998 highlighting a recent #preprint from Kylin Emhoff, @jcoker10 and Jan Joseph Melenhorst challenging the concept that MCL1 promotes chemoresistance from the nucleus. prelights.biologists.com/highlights/mcl…

We are happy to highlight a new preprint on @biorxivpreprint where we reveal a sex-specific estrogen dependent role for platelet PAR4 signaling in regulating CD8+ T cell immunity in glioblastoma. doi.org/10.1101/2025.0… @JusinLathia @2Scottish @CCLRI

Our WDR5 journey in #GBM continues, happy to highlight a new @biorxivpreprint on a new series of inhibitors in collaboration with the @CCLRI team - @SStaufferPhD @jcoker10 @SamuelSprowls @AnthonyrSloan biorxiv.org/content/10.110…

🚀 On July 21, the Chemical Probes Portal celebrates 10 years! Launched after the landmark 2015 @NatChemBio paper by Arrowsmith et al., it now features >1150 compounds, >600 targets, and 1600+ expert reviews—powering better biomedical research. 🧪🔬 👉 chemicalprobes.org

Breaking news: Staff at the National Institutes of Health have issued a declaration calling on their director to depoliticize the agency and reverse spending cuts. scim.ag/3FN7HHK

This is actually so embarrassing. How do fans have more respect for the game than the league/broadcasters??? These playoffs have been all-time awesome and this is what the “minds” at ESPN give us. I’m so annoyed 😂

The NBA and ESPN have responded to complaints about the lack of pomp and circumstance for The Finals by CGI ADDING A TINY BLURRY TROPHY ONTO THE COURT. Give me a break 🤦‍♂️

Spike Lee is the kind of rich person I would be. No yachts, no grandiose political power trips, no corporate raiding. Just get me courtside to every important sporting event on earth!