Sequence design at the upcoming Kipoi webinar on Wed :-) Looking forward to Sager Gosai's talk!

Join us for an exciting afternoon of RNA talks (including Giulia Cantini @[email protected] from my lab) on the 10th of July at MPI for Biochemistry! We will celebrate summer with some pizza and drinks afterwards Computational Health Center MUDS

We explored why pathogenic variants are present in the gnomAD database for early-onset, severe, dominant conditions. Explanations identified for most but some likely due to incomplete penetrance though more study needed. 1/3

if you are passionate about the therapeutics of the future and want to be part of the 'RNA revolution' apply for this project in our lab: we want leverage the latest AI models for the prediction of potential small molecules targeting RNAs in the cell! t.ly/BjO5o

A nice example of how training set design can have huge impact on generalizability. This continues to be an underappreciated and underutilized trick for improving models.

After DeepRVAT, FuncRVP is our next rare variant association testing tool! Rather than testing one gene at a time, we leverage information across genes using foundation models of gene function. All open source. Shubhankar Londhe Shubhankar Londhe led this cool work - more in his thread

Thrilled to share our work on nucleotide dependencies. We can see the base relationships DNA language models capture. TF motifs and their interactions, splicing, RNA structures...the breadth of biology unraveled is awesome. And it matters for variant effect prediction! Enjoy.

New concept for variant effect prediction. 'How bad' is the new 'how rare'. With DNA LMs we can ask how important a variant is for predicting other bases. This influence score correlates better with functional assays. This was not possible with alignment-based conservation.

Join us for the next Kipoi Seminar ft. Max Horlbeck & Ruochi Zhang (Ruochi Zhang) from Jason Buenrostro's lab at Harvard University & Broad Institute! 👉Single-cell footprinting reveals the organization of cis-regulatory elements 📆Wed Sep 4, 5:30pm CET 🧬tum-conf.zoom.us/meeting/regist…

Proud to be part of the European Rare Diseases Research Alliance ERDERA. Looking forward to fruitful collaboration across Europe to accelerate and increase the rate of RD diagnostics.

Taking sequence-based modeling to the next level: 500 kb DNA input, single-cell resolution, joint ATAC & RNA coverage tracks. Thumbs up to Johannes Johannes Hingerl & Laura Laura Martens (@lauradmartens.bsky.social). Wonderful collab with Jason Buenrostro & Fabian Theis. Open source, of course. Enjoy!

Really pleased to see DeepRVAT published. A new method for rare variant genetics by Eva Holtkamp & Brian Clarke in collab with gagneurlab. I like to think of DeepRVAT as one network on top of another, harnessing variant annotations from bags of models- AlphaMissense and alike.

OBS #Genomics Seminar 📆 Thursday, 9/26, 11 am ET/5 pm CET 🎙️ Miquel Anglada Girotto Miquel Anglada-Girotto (m1quelag@bluesky) Centre for Genomic Regulation (CRG) serranoLab 🎙️ Xavier Hernandez Alias Xavier Hernandez Alias Max Planck Institute of Biochemistry (MPIB) gagneurlab 👉 Registration: us02web.zoom.us/meeting/regist… #openscience Moderator: Anna Salamero Boix

DeepRVAT now out! High potential for rare disease res.: a) applies to very imbalanced case-control settings, b) our gene impairment score is trained w/o the pervasive disease-gene list bias: the strength of training on UKBB. Gr8t work Brian Clarke Eva Holtkamp w Oliver Stegle

Avantika Lal @lal_avantika today at the Kipoi seminar - modeling single-cell gene expression in health and disease from sequence. Very much looking forward!

Honoured to have written this review with Jernej Ule’s lab on the splicing code, spanning modeling, mechanisms and its implication for therapeutics designs. Wonderful teamwork.

Information leakage due to homology: A pervasive and non-trivial issue in bioinformatics. Excited to hear about Abdul Muntakim Rafi's insights and recommendations for sequence-based modeling on Wednesday at the Kipoi seminar.

Tomorrow at #ASHG2024, Shubhankar Shubhankar Londhe will present FuncRVP, our model improving rare variant testing by integrating gene impairment scores from DeepRVAT with functional gene information. Don’t miss it: we got fresh results since our July preprint! doi.org/10.1101/2024.0…

Starting in 15 min!

We moved to Bluesky: gagneurlab.bsky.social