SPLDExtraTrees: robust machine learning approach for predicting kinase inhibitor resistance

Abstract

Drug resistance is a major threat to the global health and a significant concern throughout the clinical treatment of diseases and drug development. The mutation in proteins that is related to drug binding is a common cause for adaptive drug resistance. Therefore, quantitative estimations of how mutations would affect the interaction between a drug and the target protein would be of vital significance for the drug development and the clinical practice. Computational methods that rely on molecular dynamics simulations, Rosetta protocols, as well as machine learning methods have been proven to be capable of predicting ligand affinity changes upon protein mutation. However, the severely limited sample size and heavy noise induced overfitting and generalization issues have impeded wide adoption of machine learning for studying drug resistance. In this paper, we propose a robust machine learning method, termed SPLDExtraTrees, which can accurately predict ligand binding affinity changes upon protein mutation and identify resistance-causing mutations. Especially, the proposed method ranks training data following a specific scheme that starts with easy-to-learn samples and gradually incorporates harder and diverse samples into the training, and then iterates between sample weight recalculations and model updates. In addition, we calculate additional physics-based structural features to provide the machine learning model with the valuable domain knowledge on proteins for these data-limited predictive tasks. The experiments substantiate the capability of the proposed method for predicting kinase inhibitor resistance under three scenarios and achieve predictive accuracy comparable with that of molecular dynamics and Rosetta methods with much less computational costs.

MdrDB: Mutation-induced drug resistance DataBase

“…In addition to basic information regarding the proteins, drugs, mutations and changed affinity, and structural data for wild type and mutant complexes, 146 calculated biochemical features and extra annotations are also provided. These features and annotations were chosen to be convenient to use for model training and sample splitting in machine learning algorithms for drug resistance prediction [40]. In addition, MdrDB is also the first database that includes drug resistance related mutation types beyond substitution mutations, and the availability of wider and more complex mutation types can be used to test the generalization of machine learning models.…”

KBPRNA: A novel method integrating bulk RNA-seq data and LINCS-L1000 gene signatures to predict kinase activity based on machine learning

“…With the maturity of high-throughput sequencing technology, proteomics sequencing data could be exploited to compute corresponding kinases’ activities [10]. Multiple computational tools which could be applied to predict kinase inhibitor resistance and selectivity have already been developed [11, 12]. Emilio Fenoy et al developed a generic deep convolutional neural network framework called NetPhosPan to kinase phosphorylation prediction [13].…”