Table 2.

Computational tools for monotherapy prediction.

Name	Availability	Purpose	Methodology and features	Reference
HNMDRP	Matlab and R code	Drug response prediction in CCLs	Genomic and compound features combined with drug–target interaction and PPI	37 Source code: https://github.com/USTC-HIlab/HNMDRP
KRL	Python code	Drug prioritization (ranking) in CCLs transferable to patients	Kernelized rank learning using genomic features, (predominantly gene expression)	117 Source code: https://github.com/BorgwardtLab/Kernelized-Rank-Learning
CDRscan	Web Applicationa	Drug response prediction in CCLs	Deep neural network trained on somatic mutations and drug compound fingerprints	46
Dr.VAE	Python code	Drug response prediction in CCLs	Semi-supervised Variational Autoencoder of gene expression that incorporates drug perturbation effects	48 Source code: https://github.com/rampasek/DrVAE
CancerDP	Web Application	Drug response prediction in CCLs	SVM models using (combination of) genomic features (mutations, CNVs, expression levels)	114 Webserver: http://crdd.osdd.net/raghava/cancerdp/
BMTMKL	Matlab and R code	Drug response prediction in CCLs	Bayesian multiview (original genomic modalities + aggregated views) multitask model	29 Source code: https://github.com/mehmetgonen/bmtmkl

A non-exhaustive summary of the most recent monotherapy prediction methods with an available web service or source code.

^aA web application has been promised by the authors, but no official implementation yet as of February 2020.