Drug sensitivity prediction with normal inverse Gaussian shrinkage informed by external data

In precision medicine, a common problem is drug sensitivity prediction from cancer tissue cell lines. These types of problems entail modelling multivariate drug responses on high-dimensional molecular feature sets in typically >1000 cell lines. The dimensions of the problem require specialised models and estimation methods. In addition, external information on both the drugs and the features is often available. We propose to model the drug responses through a linear regression with shrinkage enforced through a normal inverse Gaussian prior. We let the prior depend on the external information, and estimate the model and external information dependence in an empirical-variational Bayes framework. We demonstrate the usefulness of this model in both a simulated setting and in the publicly available Genomics of Drug Sensitivity in Cancer data.

[1]  Lodewyk F. A. Wessels,et al.  TANDEM: a two-stage approach to maximize interpretability of drug response models based on multiple molecular data types , 2016, Bioinform..

[2]  Nci Dream Community.  A community effort to assess and improve drug sensitivity prediction algorithms , 2014 .

[3]  Tero Aittokallio,et al.  Machine learning and feature selection for drug response prediction in precision oncology applications , 2018, Biophysical Reviews.

[4]  Francisco Azuaje,et al.  Computational models for predicting drug responses in cancer research , 2016, Briefings Bioinform..

[5]  G. G. Galli,et al.  The landscape of cancer cell line metabolism , 2019, Nature Medicine.

[6]  G. Casella.  Empirical Bayes Gibbs sampling. , 2001, Biostatistics.

[7]  Susumu Goto,et al.  KEGG: Kyoto Encyclopedia of Genes and Genomes , 2000, Nucleic Acids Res..

[8]  O. Barndorff-Nielsen.  Normal Inverse Gaussian Distributions and Stochastic Volatility Modelling , 1997 .

[9]  Zhi Zhao,et al.  Structured penalized regression for drug sensitivity prediction , 2019, Journal of the Royal Statistical Society: Series C (Applied Statistics).

[10]  On the specification of prior distributions for variance components in disease mapping models , 2016 .

[11]  Gwenaël G R Leday,et al.  An empirical Bayes approach to network recovery using external knowledge , 2016, Biometrical journal. Biometrische Zeitschrift.

[12]  James G. Scott,et al.  Shrink Globally, Act Locally: Sparse Bayesian Regularization and Prediction , 2022 .

[13]  Tero Aittokallio,et al.  Drug response prediction by inferring pathway-response associations with kernelized Bayesian matrix factorization , 2016, Bioinform..

[14]  A. Gelman.  Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper) , 2004 .

[15]  Magnus M Münch,et al.  Learning from a lot: Empirical Bayes for high‐dimensional model‐based prediction , 2018, Scandinavian journal of statistics, theory and applications.

[16]  L. Wasserman,et al.  The Selection of Prior Distributions by Formal Rules , 1996 .

[17]  Jukka Corander,et al.  Composite local low-rank structure in learning drug sensitivity. , 2019, 1905.00095.

[18]  Juan Pablo Lewinger,et al.  Incorporating Prior Knowledge into Regularized Regression , 2020, bioRxiv.

[19]  R. Tibshirani,et al.  Varying‐Coefficient Models , 1993 .

[20]  David M. Blei,et al.  Variational Inference: A Review for Statisticians , 2016, ArXiv.

[21]  Ralf Werner,et al.  The Normal Inverse Gaussian Distribution for Synthetic CDO Pricing , 2007 .

[22]  Yang Ni,et al.  Bayesian Hierarchical Varying-Sparsity Regression Models with Application to Cancer Proteogenomics , 2018, Journal of the American Statistical Association.

[23]  S. Ramaswamy,et al.  Systematic identification of genomic markers of drug sensitivity in cancer cells , 2012, Nature.

[24]  H. Jeffreys.  An invariant form for the prior probability in estimation problems , 1946, Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences.

[25]  C. Carvalho,et al.  Decoupling Shrinkage and Selection in Bayesian Linear Models: A Posterior Summary Perspective , 2014, 1408.0464.

[26]  Emanuel J. V. Gonçalves,et al.  A Landscape of Pharmacogenomic Interactions in Cancer , 2016, Cell.

[27]  R. Shoemaker.  The NCI60 human tumour cell line anticancer drug screen , 2006, Nature Reviews Cancer.

[28]  Jean-Michel Marin,et al.  Mean-field variational approximate Bayesian inference for latent variable models , 2007, Comput. Stat. Data Anal..

[29]  Geoffrey E. Hinton,et al.  A View of the Em Algorithm that Justifies Incremental, Sparse, and other Variants , 1998, Learning in Graphical Models.

[30]  Sridhar Ramaswamy,et al.  Genomics of Drug Sensitivity in Cancer (GDSC): a resource for therapeutic biomarker discovery in cancer cells , 2012, Nucleic Acids Res..

[31]  Gwenaël G R Leday,et al.  Gene Network Reconstruction using Global-Local Shrinkage Priors. , 2015, The annals of applied statistics.