Rohit Singh

Which of these images is not like the other and why does it matter for cancer research? That fourth transform is what real tissues need but most spatial methods can't handle. Enter SAME, our algorithm for integrating multimodal spatial omics across near-serial sections.

Being able to take a supervisory view, stop the LLM from trying to do something dumb, and debug where needed definitely is a huge part of what makes this possible. On the other hand, I’ve saved weeks of sheer manual typing time already.

I find that Claude Code planning mode is only useful for really small tasks - for larger tasks what's working better for me: Ask Claude to make a plan in a markdown file -- then promptly take that Claude out back and shoot it (ctrl+c) Spin up a new Claude and tell it: "a…

SAME: Topology-flexible transforms enable robust integration of multimodal spatial omics biorxiv.org/content/10.110… #biorxiv_bioinfo

SAME: Spatial Alignment of Multimodal Expression: Topology-flexible transforms enable robust integration of multimodal spatial omics biorxiv.org/content/10.110…

The Duke Center of Excellence for Multiscale Immune Systems Modeling (MISM) will bring together experts from 14 departments at Duke to build advanced computational models that will improve understanding of infectious and immune-related diseases. medschool.duke.edu/news/new-hub-f…

🚨 Attention US PhDs in Bioinformatics: Interested in a postdoc in my group at Bar-Ilan University 🇮🇱 through the Fulbright Israel fellowship (2026–2027)? 🧬 2 years of research 💰 $95,000 stipend + relocation + benefits 📅 Apply by 15/9 🔗 fulbright.org.il/program/2/842 ✉️ Email me!

Next Tues (7/1) at 4PM ET, @NathanielBlalo2 and @srivsesh will present "Functional alignment of protein language models via reinforcement learning" Paper: biorxiv.org/content/10.110… This will be followed by @WengongJin, who will give an early-career talk on Wed (7/2) at 4PM ET!