PaccMann: Prediction of anticancer compound sensitivity with multi-modal attention-based neural networks

TL;DR

This paper introduces PaccMann, a multi-modal neural network that predicts anticancer drug sensitivity by integrating molecular structures, gene expression profiles, and protein interaction knowledge, utilizing attention mechanisms for improved accuracy and interpretability.

Contribution

The study develops and compares attention-based encoders for drug and gene data, demonstrating superior performance over traditional fingerprint-based models in predicting drug sensitivity.

Findings

Attention-based encoders outperform baseline models.

Model interpretability is enhanced through attention mechanisms.

Genes, bonds, and atoms important for predictions are identified.

Abstract

We present a novel approach for the prediction of anticancer compound sensitivity by means of multi-modal attention-based neural networks (PaccMann). In our approach, we integrate three key pillars of drug sensitivity, namely, the molecular structure of compounds, transcriptomic profiles of cancer cells as well as prior knowledge about interactions among proteins within cells. Our models ingest a drug-cell pair consisting of SMILES encoding of a compound and the gene expression profile of a cancer cell and predicts an IC50 sensitivity value. Gene expression profiles are encoded using an attention-based encoding mechanism that assigns high weights to the most informative genes. We present and study three encoders for SMILES string of compounds: 1) bidirectional recurrent 2) convolutional 3) attention-based encoders. We compare our devised models against a baseline model that ingests engineered fingerprints to represent the molecular structure. We demonstrate that using our attention-based encoders, we can surpass the baseline model. The use of attention-based encoders enhance interpretability and enable us to identify genes, bonds and atoms that were used by the network to make a prediction.