Publications

How did we get a regression model to the top of the Virtual Cell Challenge Our models

December 9, 2025 Updated March 20, 2026

In the Virtual Cell Challenge, using only the pseudobulks from single-cell data was enough to get most of what could be learned.


Read post

Competition write-up

Poster

A cautionary tale on biological truths Understanding data

March 13, 2026

Be very careful when adding "well-known biology" in the model as inductive bias. We learned it may backfire spectacularly.


Read post

Make multimodality work for you, not against Understanding data

February 25, 2026

Is multi-modal data the next step in understanding biology? It might be, with caveats. Thankfully, how helpful a new mode is can be tested in many ways.


Read post

Utility of Virtual Cells

Utility of Virtual Cells Perspective

January 7, 2026 Updated February 10, 2026

We've published a workflow on how to read and evaluate Virtual Cell papers to best understand their domain of applicability.


Read post

Downloadable whitepaper

What can Virtual Cells do for you today?

What can Virtual Cells do for you today? Perspective

November 25, 2025

Turns out, no matter the specific assay, today's limitations of Virtual Cells are quite uniform: they perform quite well predicting a known perturbation to unknown biosamples, they may or may not work when the perturbation itself is untrained, and they don't work at all when the entire assay is new to the model.


Read post

Masked BERT-like Pretraining of Single-cell Foundation Models Does Not Improve Virtual Cells Pretraining and foundation models

April 16, 2025 Updated November 20, 2025

Investigating the lack of performance gains from native pretraining revealed that the way we currently train native single-cell data carries much less information than expected.


View poster

Pretraining virtual cells is useless

The Patient Prediction Puzzle Understanding data

October 16, 2025

Real-world patient data is inherently biased. With the correct controls applied, little distinguishing molecular information remains in TCGA data.


Read post

Benchmarking foundation cell models for post-perturbation RNA-seq prediction Pretraining and foundation models

September 23, 2024 Updated July 25, 2025

Cell foundation models pretrained on native data add little to prediction results.


View poster

Do cell foundation models work?

Read full article

IC50 is a deep rabbit hole Understanding data

June 30, 2025

We frequently say that data harmonization is not optional in biology, it's essential. Here are some examples on how easy it is to teach the wrong idea to the machine just by feeding it unprocessed data.


Read post

Why can we predict biology? Perspective

May 22, 2025

Predicting biology is a very different problem from predicting chemistry. This is because the rules of chemistry are bottom-up and come from physics, the rules of biology are top-down and come from evolution using whatever it happened to have at its disposal.


Read post

Computational modeling of immune cell phenotypes enhances prediction of clinical anti-cancer drug response Our models

May 1, 2025

Adding predicted cytokine activity to an ensemble model may help a bit translating in vitro data to the clinic.


View poster

How to make data points worthwhile Understanding data

January 11, 2025

A theoretical treatise on efficiency gains expected from a simulation-guided lab-in-the-loop system


Read post

In defense of RNASeq Perspective

October 24, 2024

Why RNASeq could still be suitable as a good snapshot of cell state despite RNA being upstream from proteomics


Read post

The EFFECT benchmark suite: measuring cancer sensitivity prediction performance – without the bias Understanding data

October 30, 2023 Updated October 7, 2024

A point we keep belaboring is that benchmarking is not trivial. One can easily assume that simple drug sensitivity prediction with computational models is a solved problem - it is still very much not. This benchmark we published aims to provide a better prediction of a model's usefulness in biotech applications, with robust scoring metrics and application-specific train-test splits.


Article

So how do you benchmark biology?

Tool page