Broad Institute - Eric and Wendy Schmidt Center

Crunch 1

Predicting the effect of held-out single-gene perturbations

In this Crunch, we will explore how well we can predict the single-cell transcriptomic response to single-gene perturbations that were not measured and provided in the training dataset.

Crunch 2

Predicting the effect of held-out double-gene perturbations

In this Crunch, we will explore how well we can predict the single-cell transcriptomic response to double-gene perturbations that were not provided in the training dataset.

Crunch 3

Identifying combinatorial perturbations to drive adipocyte differentiation

White and brown adipocytes play distinct roles in energy balance, and identifying perturbations that shift cells toward thermogenic states could open new therapeutic opportunities for obesity and metabolic disease. Recent single-cell CRISPR studies now make it possible to measure how gene perturbations reshape adipocyte state at scale, but the key combinatorial regulators remain poorly understood.

In Crunch 3, participants are asked to nominate combinatorial gene perturbations that are predicted to most strongly increase a target transcriptional program, with particular emphasis on thermogenic potential as quantified by the objective function defined below. Crunch 3 will include two evaluations.

Challenge data will be published when available. Check back soon.

Crunch 1

First Place

Pearson-Delta

Participant(s): Laurent RENARD (trottefox)
Details: France

Obesity Machine Learning Competition: Tackling Metabolic Diseases

Crunch 1

Predicting the effect of held-out single-gene perturbations

Crunch 2

Predicting the effect of held-out double-gene perturbations

Crunch 3

Identifying combinatorial perturbations to drive adipocyte differentiation

References