Artificial Intelligence

Analyzing perturbed cell morphologies using deep learning (11/20)

A seminar by Matteo Di Bernardo, a graduate student in the Cheeseman Lab, developing computational approaches for microscopy-based functional genomics screens. He combines scalable data processing methods with machine learning to analyze cellular mechanisms through high-throughput microscopy. His work currently focuses on morphological screens that probe fundamental cell biology processes, and he aims to apply these computational tools to more complex cellular states in development and cancer.

Machine learning helps predict drugs’ favorite subcellular haunts
Biochemistry

Researchers in Whitehead Institute Member Richard Young’s lab and colleagues at MIT show that a machine learning model can predict which subcellular compartments a drug will concentrate in based on its chemical features. This could be used to design safer and more effective drugs, and to understand how subcellular compartments govern diverse biochemical processes.