PhD bioinformatics, genomics
United States
Joined August 2012
- Tweets1.2K
- Following488
- Followers226
- Likes1.4K
Vasantika retweeted
BioHub launches the Virtual Biology Initiative: An open resource of AI-ready biological data, as a foundation for AI models that ultimately cure all human disease. HCA is pleased to work alongside this effort that includes scientific communities to maximize its impact.
🧵 Biohub, @ENERGY, @NIH, and new funding partners today announced a major expansion of an international effort to generate and make accessible the data enabling predictive AI models of biology. Together, the organizations are investing $1.8 billion in funding, data, computation and new measurement technology - largest coordinated commitment to map the biology of the cell to power development of digital models of biology. bit.ly/3UhjoxD
Vasantika retweeted
I’m hiring a postdoc at Cambridge again!
Work on privacy-preserving human pangenomes at the intersection of statistics, graph algorithms, optimization, differential privacy & computational genomics.
Please spread the word!
jobs.cam.ac.uk/job/57231/
Vasantika retweeted
Remarkable Magnum Opus from @VamsiMootha & the Mitocarta Tree of Life consortium consisting of 9 back to back research papers (+2 perspectives) just released in Cell and Cell family journals:
- Characterization of mito proteomes from 6 diverse species, from Giardia to Leishmania to Arabidopsis
- Optimized methods for mito enrichment, DIA mass spec, protein correlation profiling, microscopy validation, and cryoEM analysis of native complexes
- Comparative analysis reveals dozens of proteins shared among pathogens but lost in human hosts = “pan-parasite” drug targets
- Data used to re-train ML predictor of subcellular localization => predicted mitoproteomes across hundreds of eukaryotes
- Giardia solute carrier discovered with functional validation
- Dynamic rewiring of Acanthamoeba mito proteome from aerobic to anaerobic energy production characterized
And tons more… The papers are instant landmarks and raise even more questions than they answer, paving the way for many future investigations of this fascinating organelle #mitochondria
cell.com/consortium/mitocart…
Vasantika retweeted
When do biological reasoning models use their biological inputs?
High performance can create a mirage of reliance on biological representations. Reasoning traces don’t always reflect the evidence.
Our paper examines these limits and paths to better biological reasoning models.
Vasantika retweeted
Well, I’d add one more with myself, as one with a PhD in biology and another PhD in AI and now working in both, to make n=2. Generating a hypothetical genome blueprint of a virus from LLM versus generating a true virus are completely different things. It is either an intentional attention-harnessing lie or true ignorance. The material, the manufacturing, and actual biological viability of paper-to-physical transformation is the gap you need to close. Like drawing a design of a chip or A-bomb, versus actually make one.
You have, and can add more checkpoints and guardrails all over this long chain of paper to physical transformation. Fear mongering is not well founded, and mostly emotional manipulation.
Vasantika retweeted
A first order result from this study is that perturbation effects are specific to cell states:
Rested and stimulated T-cells have distinct responses to purturbations.
==> If you build and test perturbation models, do it with the relevant cell types and cell states.
Vasantika retweeted
Open source powers the worlds infrastructure because it brings everybody up together through collaboration and transparency and openness.
Now we have the first open source community for AI tooling and data with @OpenWALDO. Come and join the community!
openwaldo.org
Vasantika retweeted
If you are working on an important bioinformatics tool, I encourage you to apply to our AI for science program for Claude credits. Link in 🧵
Vasantika retweeted
😊 I also just uploaded a new version of the lecture notes. cs229.stanford.edu/main_note… (it's still, and perhaps will always be work-in-progress, given the field is moving fast. Any mistakes are mine :) and any feedback is more than welcome :).
Instead of watching 2 hours of Netflix tonight, watch this Stanford lecture. It’s the clearest end-to-end explanation I’ve seen of how ChatGPT and Claude are actually built.
From tokenization and BPE all the way to the transformer architecture, training pipeline, and next-token decoding.
Whether you’ve never written a line of AI code or you’ve been shipping agents every day, this 2h 34min session will click things into place that most people take years to figure out.
Bookmark it and clear the time this weekend, because this might be the single most valuable thing you learn all month.
Vasantika retweeted
1/5
Cell identity is written in the proteome, not in the DNA, and not always in the RNA. Out on bioRxiv today: The first cell type-resolved, MS-based proteomic atlas of the human body.
biorxiv.org/content/10.64898…
Vasantika retweeted
🚨Today in @Cancer_Cell: 50,000 tumor genomes via MSK-IMPACT. We map which HLA alleles are highest yield for antigen discovery—and define when HLA LOH constrains targetability.
A framework for scaling TCR therapies.
@MSKCancerCenter @parkerici authors.elsevier.com/sd/arti…
Vasantika retweeted
Just updated my HHMI Gilliam Fellowship application guide for the 2026 competition, which brings major program changes.
Int'l PhD students can now apply. MD-PhDs are eligible. & a new postdoc pathway that can extend HHMI support up to 7 years total.
yaelcourtney.com/resources-a…
Vasantika retweeted
Interesting perspective by @SylvieDelacroix on how AI may reshape professional judgement. Designing AI for "collective learning" could be critical as we integrate these tools into science and healthcare.
nature.com/articles/d41586-0…
#AI #AIethics #Healthcare
Vasantika retweeted
GIP results are out. The beauty of human diversity in India hasn't faded a bit. Out of the 5000-odd communities, the largest genetic dataset for India includes only about 80 pops. This means these results are just 1.6% of the whole story 1/n #india
medrxiv.org/content/10.64898…
Vasantika retweeted
I spun up a fresh claude code session to test this! I fed my coding agent the original dataset that made us think something was off and monitored to see how much prompting the agent would need before it became similarly suspicious. (1/n)
Vasantika retweeted
Interestingly, across this cohort and gnomAD (N >1.4M), we now have evidence of homozygote carriers of pLoF variants (human knockouts) for ca. 8,000 genes (out of ca. 20,000 in the human genome)
Vasantika retweeted
Startups are providing tantalizing glimpses of AI-assisted antibody discovery and claiming game-changing potential, even as new models remain hidden and industry awaits full disclosure and benchmarking nature.com/articles/s41587-0…
[🔓read free with login]
Vasantika retweeted
Long reads carry multiple small vars and SVs and their phasing. LongcallD is the only caller that tightly integrates germline/mosaic small/structural vars/MEIs and their phasing in a single C program. One command line to get competitive small variant calls and better SVs.
New preprint on longcallD: a unified framework for joint calling and phasing of small, structural and mosaic variants from long reads. Improved SV calling and competitive small variant calling. Supervised by @lh3lh3 , co-work with @QianAlvinQin1, @wwliao88 and @irahall9.