Computational Biologist, Assistant Professor of Statistics; @cancer_dynamics @Columbia; AI/Stat ML for Bio; AI for self-organizing systems
New York, USA
- Tweets 199
- Following 557
- Followers 1,118
- Likes 1,875
Pinned Tweet
My grad student @zgr2788 and I are very excited to share our new work, Differentiable Design for Morphogenesis biorxiv.org/content/10.64898… (waxMorph)
2
10
59
6,540
Really appreciated this piece by @lpachter . A good time to reconsider how we teach and mentor, how we share ideas kindly and respectfully and how we build stronger communities both in silico and in IRL.
I've considered signing the "Severe Misalignment of AI in Mathematics" declaration because I care about the future of mathematics. But I'm not going to. Here's why
liorpachter.wordpress.com/20…
5
390
We need a few more reviewers for MLCB 2026. See @pkoo562 's message below. DM me or him with your contact email ❤️
[Reviewer Request] I need a few emergency reviewers for MLCB 2026. Please let me know if you have capacity to help over the next day or two.
DM me with your contact email!
Please RT!
1
2
645
Science grows through conversation and by widening who gets to shape it. Join us in Sydney to explore the opportunities and roadblocks for AI in Bio, and how we build foundations for knowledge that is transparent, diverse, and robust.
🧬 We are excited to announce the @NeurIPSConf 2026 workshop, I Can’t Believe It’s Not Better: Failure Modes of AI in Biology (ICBINB-BIO), to be held in Sydney, Australia, on Dec. 11 or 12, 2026!
📣 We welcome full (8 pages) and tiny (4 pages) papers that share unexpected challenges in developing, evaluating, and deploying models on biological tasks and data. Sharing these experiences helps the community avoid unproductive paths, clarify failure modes and boundary conditions, and build a culture of transparency. We will also have an exciting lineup of speakers sharing insights from the frontiers of AI and biology.
🌿 ICBINB-BIO is the official biology branch of the broader @ICBINBWorkshop initiative, united by a commitment to candid failure analysis, negative results, unexpectedly strong simple baselines, and benchmarks that genuinely support the claims made from them.
📅 Key Dates (AoE)
Submission Deadline: Aug. 29, 2026
Acceptance Notification: Sep. 29, 2026
🌐 Additional details: icbinb-bio.github.io/
📄 Submission link: openreview.net/group?id=Neur…
We also welcome you to sign up as a reviewer to evaluate submitted manuscripts and provide valuable comments and feedback as part of the selection process: forms.gle/c7bs7nLWm6s3SSkW7
We hope you can join us in Sydney to help shape a more transparent and reliable future for AI in biology! ✨
1
4
409
Very grateful for the opportunity to talk about our work on AI for Bio to understand morphogenesis and self-organizing systems @ISSCR Looking forward to all the great conversations over the next few days! Great food too!
6
374
My grad student @zgr2788 and I are very excited to share our new work, Differentiable Design for Morphogenesis biorxiv.org/content/10.64898… (waxMorph)
2
10
59
6,540
We built waxMorph with spatial omics and light sheet microscopy in mind. In the future, we hope to work with time resolved measurements to quantify how these aspects of cell biology interact across scales to shape tissues over time.
Reach out if you have such data already!!!
1
6
286
Oh and a really cool thing is that we used NVIDIA warp! github.com/NVIDIA/warp thanks @nvidia for making it available to the community so we can build open source tools to advance open science!
5
232
How are two embryos alike?
As we collect spatio-temporal microscopy data, we want to quantify variability in the timing of key developmental events.
Alignment of multiple videos is a core engineering challenge here and we have a solution; read about it: biorxiv.org/content/10.64898…
6
23
1,337
Excited for the chance to present our framework for interpretable disentanglement in multi condition data at ICML' 26 arxiv.org/abs/2506.17182 . While useful for different types of data, we think it's particularly useful if you have omics collected across distinct covariates!
5
20
1,431
bianca dumitrascu retweeted
3D genome architecture underlies health & disease, but its biochemical basis is hard to study at scale. We present Plate-C: a screening platform that profiles thousands of whole-genome 3D maps in a day ($4 each), discovering many pathways that rewire DNA: biorxiv.org/content/10.64898…
ALT Plate-C measures whole-genome architecture at scale
4
74
297
33,783
We're looking for two postdocs to join our team! Topic 1: interpretable diffusion models for multi-scale morphology + genomics data! Topic 2: generation and control of biomaterials! If interested, send me CV + 3 recommendation letters over email!
6
17
1,220
"As most of us learn in primary school, a key reason authors cite relevant prior work is to help establish support for claims they make. If we don’t necessarily require that those references link to real research, then what are we even doing? "
Machine learning research is not serious research and therefore hallucinated references are not necessarily a big deal, agrees a prestigious group of machine learning researchers
statmodeling.stat.columbia.e…
2
11
1,918
bianca dumitrascu retweeted
Excited for the AI & Biology Conference in Suzhou, China (Apr 20-24, 2026)! Don't miss the abstract deadline: Feb 13, 2026. Submit your work on Genome x AI, Protein x AI & more—plus fellowships up to $500. Register now: csh-asia.org/?content/3008
8
37
238
21,855
bianca dumitrascu retweeted
📢 Our Dept. of Systems Biology @Columbia has an open tenure-track Assistant Professor position in the broad area of quantitative biology. Come join our awesome department in NYC! Please circulate.
Apply: apply.interfolio.com/177622
Suggested deadline: Dec 15, 2025. @ColumbiaSysBio
45
95
17,578
bianca dumitrascu retweeted
🔬Dr. Bianca Dumitrascu led a new study published in
@NatureMethods on a computational pipeline linking spatial transcriptomics + mechanical forces in tissue development. Read more: bit.ly/4kCgT1q #CancerResearch #Mechanobiology #SpatialTranscriptomics #IICDColumbia
ALT An overview of the spatial mechano-transcriptomics pipeline (Dumitrascu Lab).
7
22
2,869
Congratulations!
Excited to announce that my master's thesis project @Cambridge_Uni has just been published in @naturemethods as a (co)-first author manuscript! We describe a computational framework to explore coordination between mechanical forces and spatial gene expression. A brief 🧵:
2
607