Soham Mukhopadhyay

Thank you sincerely for this honour 👏 I am deeply grateful and moved because this award comes from my home, l' @universitelaval , but this award is not only about me. It is first and foremost about my team, everyone who has been part of the lab, past and present: more than 80 undergraduate, graduate, and PhD students, postdoctoral fellows, research professionals, and visiting scholars who have helped build what we are today. Thank you to my department, my faculty, and to all those who have supported this journey, from Cuba to Mexico, from the United States to Canada, and to my collaborators around the world. Thank you to the funding agencies that have believed in our ideas and our passion, and to the agricultural community in Québec and across Canada, with whom we work every day. In the end, the EdeLab serves one purpose: contributing to a better society and a more sustainable environment. Thank you 💚 @NSERC_CRSNG , @canolacouncil, @CanolaGrowers, @AlbertaCanola, @SaskOilseeds, @westerngrains, @RqradQc, @CIFAR_News, @Genomequebec, @InnovationCA, @ulcriv News: services-recherche.ulaval.ca/prix-et-distin…

😂😂

👀

New work out in Nature now shows all of the nucleobases of our DNA And RNA being found in asteroid Ryugu as well! While nucleobases aren't the same thing as nucleosides or nucleotides and this finding in no way suggests that DNA or RNA molecules exist beyond Earth, it's still incredible to keep finding more of the basic building blocks of life as we know it well beyond our home. nature.com/articles/s4155…

🚨 New from the lab! In this preprint, we show how the clubroot pathogen Plasmodiophora brassicae uses a PBS3-like effector (PbGH3) to hijack plant hormone signalling and reshape root development during infection 👏 We found that PbGH3 acts as a structural and functional mimic of the immune enzyme AtPBS3 (GH3.12), evidence of the evolutionary relationship between the clubroot pathogen and Brassicas 🥦 This study is now under review 🤞

@OdedRechavi @biorxivpreprint @Edel_PLopez 😂

What's the German word for the feeling of waiting for your @biorxivpreprint preprint to be online?

Duke et al. (2026-03 preprint, bioRxiv)（オープンアクセス open access） 「完全に年代付けされた完全な生命の樹の構築」 Assembling a fully-dated complete tree of life

Excited to share this preprint that describes my latest work on using GPUs to accelerate processing of RNA-seq data. The title says it all: "RNA-seq analysis in seconds using GPUs" now on biorxiv biorxiv.org/content/10.648… Figure 1 shows they key result

The emerging role of kinase fusion proteins in cereal immunity. nature.com/articles/s4158…

@ancornman1 That's fast! 👍 Can it take two FASTA files as input and screen one against the other, or is it all-vs-all for now? Secondly, what is the minimum GPU power it needs - A100 level or a local RTX 4000 series? Thanks!

Predicting protein-protein interactions (PPIs) at proteome scale can take months with co-folding models due to massive all-vs-all comparisons required. We are excited to announce FlashPPI, a contrastive model that predicts proteome wide physical interfaces in minutes. 1/🧵

👀

A regulatory network promotes apoplastic alkalinization to prime plant immunity in tissues distal to site of infection cell.com/cell/fulltext/…

The future is complete genomes. Is it now time to move from variant calling to directly use genome assemblies? Assemblies 10x more accurate than SOTA var calling. Impressive! Advances in pangenomes / assembly pipeline opening a new way of working in genomics. #AGBTGM26

I tested models from @AnthropicAI @OpenAI @Google @Zai_org @MiniMax_AI and @Kimi_Moonshot on whether they can create a publication-level view of a protein-ligand binding site and the results were surprising. TL;DR: Anthropic models did the best and Gemini models did the worst. Task was simple: "Load 5DEL and create a publication ready view of the ORO and FMN binding site" Here are the results: 1. Sonnet 4.6 from @AnthropicAI: - Fast - Used the tools pretty well - Created the best view with clear labelling and color choices - always felt in control of what it was going for - very impressed

Introducing Clair3 v2 - signal-aware variant calling for ONT long reads. It uses move table dwelling times to dramatically boost indel detection performance. Also now fully migrated to using PyTorch for better usability. (1/) 🔗github.com/HKU-BAL/Clair3 📄biorxiv.org/content/10.648…

@spottedinprod @acmeweather feels like Anthropic made a weather app!

acme · dark sky team is back by @acmeweather

In 2000, the first human genome sequence cost between $500M & $1B In 2006, the cost was $20M Two years ago, the cost is around $600. Now it is $100.

We have published a methods paper describing how to identify cell surface receptor complexes by immunoprecipitation and map their phosphorylation sites. We hope this protocol will be useful to the community. link.springer.com/protocol/10.10…

I'm really impressed that Anthropic is explicitly focusing on life science researchers in their evals. It's great to see more attention to researcher productivity! (this figure from the Opus 4.6 system card)

Excited to join this awesome team as Associate Editor @MPMIjournal, looking forward to engaging with the community.