Senior Scientist @argonne and @UChicago. Orcid: 0000-0003-2130-2887. likes=bookmarks

Chicago,IL
Joined October 2007
Ravi Madduri retweeted
New preprint: teaching single-cell foundation models to learn from perturbations in collaboration with @madduri & @RUNX1Research supported by @cziscience. Closing the loop identified a target for RUNX1-FPD now in Ph2 trials 🔗biorxiv.org/content/10.1101/… 🧵by @yash_persh👇
We are excited to share our work on "closing the loop" 🔄 - a framework enabling single-cell foundation models (scFMs) to learn from experimental validation and dramatically improve their predictive accuracy! (1/n)
8
1
39
4,905
🧪 Now published in Bioinformatics Advances: “Benchmarking accelerated next generation sequence (NGS) analysis pipelines”   Read the full paper here: doi.org/10.1093/bioadv/vbaf0… Authors include: @madduri, @A_D_Rowe
1
2
12
658
Interact with all of the results in the Pleiotropy Atlas (PLATLAS)! platlas.cels.anl.gov/ [2/3]
1
2
5
911
Very excited to share our preprint led by @MGLevin @skoyamamd @jakob_woerner & with @damrauer assessing genome-wide pleiotropy of >1,000 clinical traits across ~1.7M individuals with nearly 30K locus-trait associations! medrxiv.org/content/10.1101/… @medrxivpreprint [1/3]
3
16
82
8,664
Great talk by Zilinghan at #AMIA on our joint work with @_anuragverma at UPenn on using LLMs in the context of rare disease research
108
The Million Veteran Program is the @DeptVetAffairs' gift that keeps on giving...critical clues to the etiology and treatment of coronary disease, the #1 cause of death worldwide. Thank you U.S. Veterans and program leadership for making this possible. Let's keep it going!
"Among all the uncertainty, you can always count on awesome science to make you smile. Check out our latest @CellCellPress interdisciplinary endeavor—a reminder that we all have a 🫀, whether right-dominant or not! sciencedirect.com/science/ar…
1
6
18
1,197
How decoding the genome can help cure disease Don’t miss today’s entry in the @AI_and_Science Schmidt Fellows Speaker Series! Featuring Hae Kyung Im (@hakyim), Associate Professor of Medicine and Human Genetics at @UChicago! datascience.uchicago.edu/eve…
1
3
255
Our paper on "Genome-wide association study of prostate-specific antigen levels in 392,522 men identifies new loci and improves prediction across ancestry groups" with @johnwitte and other collaborators has been accepted for publication in @NatureGenet !
1
120
check out our new preprint led by @Charles_Zhou12 and supervised by @MengjieChen6 and me doi.org/10.1101/2024.11.11.6… where we present scPrediXcan which integrates deep learning and single cell expression data into a powerful cell type specific TWAS framework
4
26
1
83
5,799
If you are at #ASHG24, we have several more presentation today 1. Platform talk on "Genome-Wide Assessment of Pleiotropy Across >1000 Traits Among 1.5 Million Participants of Diverse Biobanks" by @MGLevin at 10:30AM FourSeason BR 2&3 @skoyamamd @damrauer @pnatarajanmd @madduri
1
3
1
10
3,733
Our group has contributed to three platform talks and four posters at #ASHG24 this year applying #HPC and #AI to Human Genetics! I am proud of all we were able to accomplish this year. cc @_anuragverma @hakyim
1
3
273
eppro02.ativ.me/appinfo.php?… #ASHG24 Please come and hear about our work on scPrediXcan: leveraging single-cell data for transcriptome-wide association studies at cell-type level through transfer learning with @hakyim in Machine Learning and AI Applications in Human Genetics session.
2
5
359
Ravi Madduri retweeted
Very much enjoyed updating the @cziscience community on work led by @yash_pershad & @alyssa_c_parker with @madduri @LabObeng @LabAgarwal @RUNX1Research to move from single cell sequencing to RUNX1 therapies at the annual CZI Cell Science meeting!
2
34
1,449
🚀 We’re excited to announce the 1.0 release of APPFL, available now on GitHub and PyPi! Key highlights include improved security features, enhanced scalability, and new tools for creating end-to-end secure federations. Read the full release notes appfl.ai/en/latest/changelog…. (1/3)
1
2
71
A huge thanks to the @doescience for their funding and support! Hear from Under Secretary for Science and Technology, Geri Richmond, on the impact of this technology: Watch youtube.com/clip/UgkxToaUid4….
1
1
67
APPFL is open-source—try it today and help advance science with secure federations! 🛡️🧪 #FederatedLearning #OpenSource #APPFL @argonne
50
Thank you to the whole team for the wonderful collaboration! Javier Gómez Ortega, @Sid_Mahesh_12, Tarak Nandi, @madduri, and @PelkaLab! nitter.cf/TheodorisLab/status/18…
We are excited to share work led by @hanimal725 and @madhavanvvs developing a quantized multitask learning strategy for network biology, built on a foundation model pretrained on ~95M single-cell transcriptomes. Manuscript: tinyurl.com/qmtl-gf Model: huggingface.co/ctheodoris/Ge…
1
1
2
632
We are one of the three!
Argonne will support three innovative artificial intelligence (AI)-driven science projects as part of the first round of awards from the National Artificial Intelligence Research Resource (NAIRR) Pilot - bit.ly/45QbhcJ
2
87
Way to go @pnatarajanmd
My conversation with @pnatarajanmd on preventing cardiovascular disease is the new Ground Truths podcast —how low to go for LDL cholesterol? — apoB, Lp(a), polygenic risk score, CHIP, role of inflammation —new RNA drugs [sorry, link is in my profile due to X-suppression]
2
144