Title:	Prediction of Single Cell Drug Response for Precision Cancer Medicine using Foundational Deep Learning Models
DNr:	NAISS 2024/6-419
Project Type:	NAISS Medium Storage
Principal Investigator:	Kasper Karlsson <kasper.karlsson@ki.se>
Affiliation:	Karolinska Institutet
Duration:	2024-12-20 – 2025-07-01
Classification:	10203 10610
Keywords:

Abstract

In the projects that run on Tetralith we will use single cell and cell barcode data to quantify tumor hetergeneity, analyze single cell data and assess tumor subclone frequency changes during drug treatment or as the primary tumor develops metastases in vivo. With cell barcode data we can trace a large number of tumor clones and see if they are sensitive or resistant to specific drugs. We will use this to find drugs that complement standard of care in different cancer types, a concept that I have termed “Precision Lethality”. In the projects that will run on Alvis we will mainly develop deep learning models for single cell drug response prediction. This is an alternative way to identify drugs to complement standard of care, and can be seen as the next generation of the precision lethality concept. We also plan to use deep learning models for in silico perturbation experiments.