Machine learning applied to omics data

TL;DR

The use of Random Forest and Penalized Multinomial Logistic Regression for integrative analysis of genomics and immunomics in pancreatic cancer is reviewed and the use of association rules with predictive purposes to overcome the low predictive power of the previously mentioned models is proposed.

Abstract

In this chapter we illustrate the use of some Machine Learning techniques in the context of omics data. More precisely, we review and evaluate the use of Random Forest and Penalized Multinomial Logistic Regression for integrative analysis of genomics and immunomics in pancreatic cancer. Furthermore, we propose the use of association rules with predictive purposes to overcome the low predictive power of the previously mentioned models. Finally, we apply the reviewed methods to a real data set from TCGA made of 107 tumoral pancreatic samples and 117,486 germline SNPs, showing the good performance of the proposed methods to predict the immunological infiltration in pancreatic cancer.

Figures (4)

• Figure 1: Definition and Matrix Data Types. (A) Genomics and (B) Immunomics.
• Figure 2: Random Forest illustration.
• Figure 3: Pairwise AUC obtained from the OOB observations for the different configurations of the RF.
• Figure 4: Pairwise AUC obtained from $k$-CV for the different configurations of the LASSO MLR.