Srinivasan Sivanandan

This paper introduces a pooled CRISPR screening platform combined with self-supervised deep learning to infer gene function from morphological phenotypes without target-specific biomarkers.

The paper uses deep learning on images of iPSC-derived motor neurons with fALS mutations to identify disease-relevant phenotypes and molecular differences.

This study develops quantitative metrics and simulated datasets to evaluate subclonal reconstruction methods in tumours.