Arvind Kumar

Life will find a way!

[Kyle Farh MD PhD - Illumina AI Lab @KyleFarh]

In a saturation MPRA of the MAPT promoter, PromoterAI tracked measured variant effects, supporting its use for prioritizing pathogenic promoter variants.
https://t.co/rw2eUT8Rr5

[Kyle Farh MD PhD - Illumina AI Lab @KyleFarh]

When linear models and deep learning give the same results, orders of magnitude more training data are needed https://t.co/gcjjIjWhf3 ;

[Kyle Farh MD PhD - Illumina AI Lab @KyleFarh]

Reminder to benchmark fancy deep learning models against linear baselines, especially when training data is lacking: https://t.co/tEDS6bTUtB

[Kishore Jaganathan @kjaganatha]

We're thrilled to introduce PromoterAI — a tool for accurately identifying promoter variants that impact gene expression. 🧵 (1/)

[William J. Greenleaf @WJGreenleaf]

Our single-cell, pan-cancer analysis of open chromatin in human cancers, done in collaboration with TCGA and the Chang and Corces labs (among others including Illumina), is out in Science today. Congrats to Laksshman Sundaram and all authors! https://t.co/4whYsPvqyN

[Arvind Kumar @arvind0422]

Our team's work on Gnomix for Local ancestry inference will be presented today at #ASHG2021 from 2:30PM PT to 3:30 PM. A big thanks to #ASHG2021 for the Reviewer's Choice award!! Congrats to @_danielmas, @alexGioannidis & @HelgiHilmars!

[Alexander G. Ioannidis @alexGioannidis]

Why use Gnomix, our new local ancestry algorithm for big data?

1. Fastest, works w/ biobank scale data & whole genomes
2. Highest resolution, can find small, ancient segments
3. Most accurate
4. Performs phase error correction (& quickly)
5. Easy to use

https://t.co/RoAzHnTdT6

[Alison Q O'Neil @alisonQoneil]

Best Paper Prize at our AI for Affordable Healthcare workshop at @iclr_conf goes to "XGMix: Local-Ancestry Inference with Stacked XGBoost" - a method to diversify genetic research beyond European ancestry: https://t.co/XH23sLjuVW #AI4AH #ICLR2020 @CanonMedicalEDI @bioandemerging