Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile
Vikram Agarwal (@vagar.bsky.social)

@vagar112

Head of mRNA Platform Design Data Science @Sanofi. R&D in mRNA therapeutics using ML/DL techniques. Formerly @MITCSBPhD, @MSFTResearch, @uwgenome, & @calico.

ID: 4337019978

linkhttps://scholar.google.com/citations?user=SQLyu2wAAAAJ&hl=en calendar_today23-11-2015 19:34:04

1,1K Tweet

3,3K Followers

1,1K Following

Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

Interesting that Nature Genetics, who editorially rejected our Enformer work because it wasn't an advance, now publishes 3 papers benchmarking Enformer (because presumably it was considered an advance?)🤔

Interesting that <a href="/NatureGenet/">Nature Genetics</a>, who editorially rejected our Enformer work because it wasn't an advance, now publishes 3 papers benchmarking Enformer (because presumably it was considered an advance?)🤔
Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

Has scientific community converged on good alternative to X yet? My ethical system is feeling increasingly misaligned with this platform as they embrace more conspiracy theorists and misinformation. Suggestions welcome

Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

It boggles my mind that people are willing to spend 7x the price on Apple Vision Pro vs Meta Quest 3 which has largely the same functionality 🤯 when I talk to users of it, most aren't aware of what MQ3 has the ability to do. 🍎 good at brand marketing I'll give em that!

Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

pubs.acs.org/doi/full/10.10… My first contribution to work in a non-bio (chemistry) journal! Fantastic team Sanofi was able to detect that certain subfractions of a lipid formulation have greater potency - helpful for mRNA therapeutic applications!!

Jian Zhou (@zhou_jian) 's Twitter Profile Photo

Glad that our sequence model of promoters in human genome is now published in Science Magazine. Check out the paper for a deep dive into the sequence basis of transcription initiation at the basepair level: science.org/doi/10.1126/sc…

Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

Was an honor to write a Perspective on the evolution of the field of deep learning in genomics, contextualizing the innovative work of Kseniia Dudnyk Jian Zhou in the same issue. Congratulations to co-writer Jun Wang @Sanofi for her contribution here too! science.org/doi/10.1126/sc…

Justin Hong (@justjhong) 's Twitter Profile Photo

How can we better reveal cellular 🦠and sample🧍variation from large-scale scRNA-seq studies? We have released a new and improved deep generative model, #MrVI, in scvi-tools and on bioRxiv, accompanied by real-world use cases. A thread... 🧵1/ biorxiv.org/content/10.110…

How can we better reveal cellular 🦠and sample🧍variation from large-scale scRNA-seq studies? We have released a new and improved deep generative model, #MrVI, in scvi-tools and on bioRxiv, accompanied by real-world use cases. A thread... 🧵1/
biorxiv.org/content/10.110…
Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

academic.oup.com/bioinformatics… fantastic collaboration between our group & Ziv & Sven's teams on ML modeling work to enhance mRNA delivery via lipid nanoparticle design!!

Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

Excited to announce release of the final CodonBERT publication to model the language of codons, trained on 10M ORFs! Great collab between our team, Sizhen, Ziv Bar Joseph, and Sven Jager's team Sanofi. GitHub code available! m.genome.cshlp.org/content/early/…

Vikram Agarwal (@vagar.bsky.social) (@vagar112) 's Twitter Profile Photo

Excited to announce that our fully computational R&D team Sanofi (mRNA Center of Excellence) has two open positions (Bioinformatics Scientist and Senior/Principal Data Scientist in Computational Chemistry), details here: linkedin.com/posts/vagarwal…

Prof. Nikolai Slavov (@slavov_n) 's Twitter Profile Photo

Predicting ribosome density on messenger RNA in mammalian cells A deep convolutional neural network predicts ribosome density based on RNA sequences. The 5′ UTR, CDS, and 3′ UTR contribute ~67%, 31%, and 2% per-nucleotide predictive power of the model.

Predicting ribosome density on messenger RNA in mammalian cells

A deep convolutional neural network predicts ribosome density based on RNA sequences. 

The 5′ UTR, CDS, and 3′ UTR contribute ~67%, 31%, and 2% per-nucleotide predictive power of the model.