_version_ 1866909690310426624
author Fahsbender, Elizabeth
Andersson, Alma
Ash, Jeremy
Binder, Polina
Burkhardt, Daniel
Chang, Benjamin
Gerber, Georg K.
Gitter, Anthony
Godau, Patrick
Gupta, Ankit
Haliburton, Genevieve
He, Siyu
Ideker, Trey
Jelic, Ivana
Khan, Aly
Kim, Yang-Joon
Krishnapriyan, Aditi
Laurent, Jon M.
Liu, Tianyu
Lundberg, Emma
Mehta, Shalin B.
Moccia, Rob
Pisco, Angela Oliveira
Pollard, Katherine S.
Ramani, Suresh
Saez-Rodriguez, Julio
Senbabaoglu, Yasin
Simon, Elana
Sivanandan, Srinivasan
Stolovitzky, Gustavo
Valer, Marc
Wang, Bo
Zhang, Xikun
Zou, James
Kalantar, Katrina
author_facet Fahsbender, Elizabeth
Andersson, Alma
Ash, Jeremy
Binder, Polina
Burkhardt, Daniel
Chang, Benjamin
Gerber, Georg K.
Gitter, Anthony
Godau, Patrick
Gupta, Ankit
Haliburton, Genevieve
He, Siyu
Ideker, Trey
Jelic, Ivana
Khan, Aly
Kim, Yang-Joon
Krishnapriyan, Aditi
Laurent, Jon M.
Liu, Tianyu
Lundberg, Emma
Mehta, Shalin B.
Moccia, Rob
Pisco, Angela Oliveira
Pollard, Katherine S.
Ramani, Suresh
Saez-Rodriguez, Julio
Senbabaoglu, Yasin
Simon, Elana
Sivanandan, Srinivasan
Stolovitzky, Gustavo
Valer, Marc
Wang, Bo
Zhang, Xikun
Zou, James
Kalantar, Katrina
contents Artificial intelligence holds immense promise for transforming biology, yet a lack of standardized, cross domain, benchmarks undermines our ability to build robust, trustworthy models. Here, we present insights from a recent workshop that convened machine learning and computational biology experts across imaging, transcriptomics, proteomics, and genomics to tackle this gap. We identify major technical and systemic bottlenecks such as data heterogeneity and noise, reproducibility challenges, biases, and the fragmented ecosystem of publicly available resources and propose a set of recommendations for building benchmarking frameworks that can efficiently compare ML models of biological systems across tasks and data modalities. By promoting high quality data curation, standardized tooling, comprehensive evaluation metrics, and open, collaborative platforms, we aim to accelerate the development of robust benchmarks for AI driven Virtual Cells. These benchmarks are crucial for ensuring rigor, reproducibility, and biological relevance, and will ultimately advance the field toward integrated models that drive new discoveries, therapeutic insights, and a deeper understanding of cellular systems.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10502
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Benchmarking and Evaluation of AI Models in Biology: Outcomes and Recommendations from the CZI Virtual Cells Workshop
Fahsbender, Elizabeth
Andersson, Alma
Ash, Jeremy
Binder, Polina
Burkhardt, Daniel
Chang, Benjamin
Gerber, Georg K.
Gitter, Anthony
Godau, Patrick
Gupta, Ankit
Haliburton, Genevieve
He, Siyu
Ideker, Trey
Jelic, Ivana
Khan, Aly
Kim, Yang-Joon
Krishnapriyan, Aditi
Laurent, Jon M.
Liu, Tianyu
Lundberg, Emma
Mehta, Shalin B.
Moccia, Rob
Pisco, Angela Oliveira
Pollard, Katherine S.
Ramani, Suresh
Saez-Rodriguez, Julio
Senbabaoglu, Yasin
Simon, Elana
Sivanandan, Srinivasan
Stolovitzky, Gustavo
Valer, Marc
Wang, Bo
Zhang, Xikun
Zou, James
Kalantar, Katrina
Machine Learning
Artificial Intelligence
Artificial intelligence holds immense promise for transforming biology, yet a lack of standardized, cross domain, benchmarks undermines our ability to build robust, trustworthy models. Here, we present insights from a recent workshop that convened machine learning and computational biology experts across imaging, transcriptomics, proteomics, and genomics to tackle this gap. We identify major technical and systemic bottlenecks such as data heterogeneity and noise, reproducibility challenges, biases, and the fragmented ecosystem of publicly available resources and propose a set of recommendations for building benchmarking frameworks that can efficiently compare ML models of biological systems across tasks and data modalities. By promoting high quality data curation, standardized tooling, comprehensive evaluation metrics, and open, collaborative platforms, we aim to accelerate the development of robust benchmarks for AI driven Virtual Cells. These benchmarks are crucial for ensuring rigor, reproducibility, and biological relevance, and will ultimately advance the field toward integrated models that drive new discoveries, therapeutic insights, and a deeper understanding of cellular systems.
title Benchmarking and Evaluation of AI Models in Biology: Outcomes and Recommendations from the CZI Virtual Cells Workshop
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2507.10502