Jian Ma

Meet the 2025 ACM Fellows! Hailing from 14 countries and chosen from among ACM's global membership of over 100,000 computing professionals, these 72 honorees are cited for contributions in a wide range of computing research areas.

Please join us in congratulating the new inductees! buff.ly/Pga16R8

Thank you Anupam!

MIMYR: Generative modeling of missing tissue in spatial transcriptomics https://www.biorxiv.org/content/10.1101/2025.11.24.690239v1

TissueNarrator: Generative Modeling of Spatial Transcriptomics with Large Language Models https://www.biorxiv.org/content/10.1101/2025.11.24.690325v1

HEIMDALL: A Modular Framework for Tokenization in Single-Cell Foundation Models https://www.biorxiv.org/content/10.1101/2025.11.09.687403v1

that was me scienc'ing the other day

Check out the recent preprints in #cellbio, selected by the @prelights.bsky.social community.

The featured preLight, from @bdmaier.bsky.social,‬ highlights STEAMBOAT, a ML framework for delineation of cellular interactions from @jianma.bsky.social & team.

journals.biologists.com/jcs/article/...

Hannenhalli & Pevzner's legendary STOC-95 paper on sorting by reversals has won the 30-year Test-of-Time Award at STOC! 30 years!! A rare algorithmic gem from #compbio with lasting impact - I still use it to introduce genome rearrangements in my class. sigact.org/prizes/stoc_...

We introduce EYKTHYR, a computational method that integrates gene expression and chromatin accessibility in a spatially aware model to identify transcription factors shaping spatial gene programs.

We introduce POPARI -- an interpretable, spatially-aware factor-based model for multi-sample spatial transcriptomics, capturing spatial patterns of gene programs and how they vary across samples and conditions.

The short talk and scholarship deadlines for the @keystonesymposia.bsky.social AI in Molecular Biology meeting in September are coming up fast, May 20th. Looking forward to seeing the submissions! www.keystonesymposia.org/conferences/...

Major nuclear locales define nuclear genome organization and function beyond A and B compartments New insights into nuclear genome organization were generated by comparing genome localization relative to three major nuclear locales- nuclear lamina, nuclear speckles, and nucleoli- across several hu...

Final version of our paper @elife.bsky.social: Major nuclear locales define nuclear genome organization and function beyond A and B compartments. elifesciences.org/articles/99116

Cellular interactions span scales and shape cell states & tissue function. We introduce STEAMBOAT, a multiscale attention model for spatial omics that reveals how cells interact - from local contact to long-range signals, uncovering previously unrecognized patterns. Led by Lane Fellow Shaoheng Liang

Integrating spatial transcriptomics across platforms is hard - different gene panels, sparse data. We introduce LLOKI, using optimal transport + single-cell FM for unified ST integration. Work led by Ellie Haber @mldcmu.bsky.social & Lane Fellow Spencer Krieger.

Congratulations Xian!

Honored to be elected as a Fellow of ISCB @iscb.bsky.social ! Grateful to the community, colleagues, and students who have shaped my journey. Look forward to contributing to the next wave of advances in comp bio!

How does genetic variation affect the cellular genome-wide activity of CRISPR-Cas9 at a population scale? Excited to have contributed to this work in collab w/ Shengdar Tsai, Yong Cheng and their teams @stjuderesearch.bsky.social. The ML component, CHANGE-net, was developed by Wendy Yang in the lab

It’s been a tough few weeks. My 10yo daughter was diagnosed with a very rare, aggressive cancer called interdigitating dendritic cell sarcoma (IDCS). I’m reaching out to identify clinicians/patients who have encountered pediatric IDCS or other (non-LCH) dendritic or histiocytic sarcomas cases.

PQG Seminar | Program in Quantitative Genomics | Harvard T.H. Chan School of Public Health The goal of the PQG Seminar Series seeks to further the development and application of quantitative methods, especially for high dimensional data, as well as focus on the training of quantitative geno...

Boston friends: I'm giving a PQG seminar next Tuesday (2/11) at @harvardchanschool.bsky.social. Hope to see some of you there! hsph.harvard.edu/research/qua...

Delighted to be hosting @jianma.bsky.social as a Distinguished Speaker (Lecturer?) at Duke Cell Biology.

Great talk on a variety of foods, from Higashi and SpiceMix to Steamboat!

Also, some stuff on 3D genome.

Was asked to contribute brief commentary about my views on the present and future of AI models for regulatory genomics. Wud love for others to add their comments to this thread.

Also check out commentaries from @jianma.bsky.social, Katie Pollard, Remo Rohs & others

www.cell.com/molecular-ce...

DNALONGBENCH: A Benchmark Suite for Long-Range DNA Prediction Tasks https://www.biorxiv.org/content/10.1101/2025.01.06.631595v1

bsky.app/profile/jian...

Fred A. Kummerow, an Early Opponent of Trans Fats, Dies at 102 (Published 2017) Artificial trans fats were ruled unsafe by the Food and Drug Administration partly in response to a lawsuit that Professor Kummerow filed against the agency.

This is so inspiring! I know Fred Kummerow (www.nytimes.com/2017/06/01/s...) published a paper when he was 100 years old: pubmed.ncbi.nlm.nih.gov/28955814/

a black background with fireworks and the words happy new year 2025 ALT: a black background with fireworks and the words happy new year 2025

Happy New Year 2025! 新年快乐~ 🎉

We’re thrilled to announce the keynote speakers for #RECOMB2025! Don’t miss the opportunity to gain insights from leading experts in the field: Peer Bork, Evan Eichler, Narry Kim, Sang Yup Lee, Lior Pachter, Olga Troyanskaya, and Leslie Valiant.

📢 Stay tuned for more updates 📢

Excited about L2G, led by Wenduo Cheng. We leverage LLMs to beat genomic FMs and strong supervised baselines on a wide range of benchmarks. L2G uses cross-modal transfer (rather than vanilla fine-tuning), and neural architecture search to learn a genomic-specific embedder model.

Can we bypass the resource bottleneck of pretraining genomic Foundation Models? Our work L2G repurposes language LLMs for genomics via cross-modal transfer, matching fine-tuned genomic FMs. Kudos to Wenduo & fantastic collab w/ @atalwalkar.bsky.social. L2G, language to genome; L2G, life’s too good!

Check out this systematic benchmark of genome-wide, annotation agnostic DNALMs & strong baseline ab-initio models for biologically meaningful tasks in regulatory genomics 1/

I am so, so sorry, Rita. My deepest condolences