Domain-invariant features for mechanism of action prediction in a multi-cell-line drug screen

Motivation: As a proxy for diseased cells, a cell line cannot be thought of as a perfect model. Many diseases feature an important molecular heterogeneity. Consequently, a drug may be effective against one molecular subtype of a disease, but less so against another. To characterise drugs with respect to their effect not only on one cell line but on a consensus of several is therefore a promising strategy to streamline the drug discovery process.

Results: The contribution of this paper is twofold. First, we investigate whether we can predict MOA at the molecular level without optimisation of the MOA classes to the screen specificities. To this end, we benchmark a set of algorithms within a conventional pipeline, and evaluate their MOA prediction performance according to a statistically rigorous framework. Second, we extend this conventional pipeline to the simultaneous analysis of multiple cell lines, each with potentially different morphological baselines. For this, we propose multitask autoencoders, including a domain-adaptive model used to construct domain-invariant feature representations across cell lines. We apply these methods to a pilot screen of two triple negative breast cancer cell lines as models for two different molecular subtypes of the disease.