David van Dijk
5 Papers
David van Dijk is an academic researcher. The author has contributed to research in topics: Computer science & Biology. The author has co-authored 1 publications.
Chat about Author
Papers
BrainLM: A foundation model for brain activity recordings
Josue Ortega Caro,Antonio H. de O. Fonseca,Christopher Averill,S. A. Rizvi,Matteo Rosati,James L. Cross,Prateek Mittal,Emanuele Zappala,Daniel Levine,Rahul M. Dhodapkar,Chadi G. Abdallah,David van Dijk +11 more
TL;DR: The Brain Language Model is introduced, a foundation model for brain activity dynamics trained on 6,700 hours of fMRI recordings that serves as a powerful “lens” through which massive repositories of fMRI data can be analyzed in new ways, enabling more effective interpretation and utilization at scale.
30
Scaling Large Language Models for Next-Generation Single-Cell Analysis
S. Rizvi,Daniel Levine,Sizhuang He,David Zhang,Zhuoyang Lyu,Rayyan Darji,Chang Li,Emily Sun,David Jeong,Lawrence Zhao,Jennifer Kwan,David Braun,Brian Hafler,Jeffrey Ishizuka,R. M. Dhodapkar,Shekoofeh Azizi,Bryan Perozzi,David van Dijk +17 more
TL;DR: Researchers develop C2S-Scale, a large language model that integrates transcriptomic and textual data to predict cellular behavior, outperforming specialized single-cell models and general-purpose LLMs in tasks like perturbation response prediction and biological reasoning.
6
Perturbational single-cell profiling of patient tumors defines lineage- and context-specific programs of innate immune resistance
C. Perry,A. Frey,Y. Fei,J. Wang,Philippos Apolinario Costa,H. N. Amin,S. Ghadermarzi,D. Levine,W. Tong,A. Koda,M. Mackie,M. He,Y. Nie,Kathryn Clulo,Feriel Ouerghi,J. Wei,Therese Cordero Dumit,M. Yaskolko,Min Ding,A. Caldera,Oleksandr Kyrysyuk,L. Lum,J. W. Allen,W. Guo,A. Naqash,A. Elliot,Ari M. Vanderwalde,M. Capelletti,T. Adeyelu,S. Barry,Miya Hugaboom,A. Bacchiocchi,Harriet M. Kluger,M. Bosenberg,A. Iwasaki,D. Braun,J. Clune,David van Dijk,K. Olino,J. Ishizuka +39 more
Causal identification of single-cell experimental perturbation effects with CINEMA-OT
Mingze Dong,Bao-Dong Wang,J. Wei,Antonio H. de O. Fonseca,Curt Perry,Alexander Frey,Feriel Ouerghi,Ellen F. Foxman,Jeffrey J. Ishizuka,R. M. Dhodapkar,David van Dijk +10 more
TL;DR: A causal inference based approach to single-cell perturbation analysis, termed CINEMA-OT (Causal INdependent Effect Module Attribution + Optimal Transport), which separates confounding sources of variation from perturbations effects to obtain an optimal transport matching that reflects counterfactual cell pairs.
Cell2Sentence: Teaching Large Language Models the Language of Biology
Daniel Levine,S. A. Rizvi,Sacha Lévy,Nazreen Pallikkavaliyaveetil,Ruiming Wu,Zihe Zheng,Antonio Oliveira Fonseca,Xingyu Chen,Sina Ghadermarzi,Rahul M. Dhodapkar,David van Dijk +10 more
TL;DR: Experiments reveal that GPT-2, when fine-tuned with C2S, can generate biologically valid cells based on cell type inputs, and accurately predict cell types from cell sentences, illustrating that language models can acquire a significant understanding of single-cell biology while maintaining robust text generation capabilities.