DeepLINK: Deep learning inference using knockoffs with applications to genomics

Abstract

We propose a deep learning–based knockoffs inference framework, DeepLINK, that guarantees the false discovery rate (FDR) control in high-dimensional settings. DeepLINK is applicable to a broad class of covariate distributions described by the possibly nonlinear latent factor models. It consists of two major parts: an autoencoder network for the knockoff variable construction and a multilayer perceptron network for feature selection with the FDR control. The empirical performance of DeepLINK is investigated through extensive simulation studies, where it is shown to achieve FDR control in feature selection with both high selection power and high prediction accuracy. We also apply DeepLINK to three real data applications to demonstrate its practical utility.

Hierarchical Microcarriers Loaded with Peptide Dendrimer‐Grafted Methotrexate for Rheumatoid Arthritis Treatment

“…After grafting with MTX, the zeta potential of the dendrimer changed a little and still carries massive positive charges, and due to electrostatic interactions, the cationic polymers can remove negatively charged inflammatory cytokines and cellular debris to impede the progression of arthritic inflammation in the RA model (Figure 2c). [ 39–42 ]…”

Transcriptomic data in tumor‐adjacent normal tissues harbor prognostic information on multiple cancer types

“…Data screening steps were applied to select more reliable survival‐related genes using external datasets from ICGC via distance correlation 24,25 . Distance correlation measures linear and nonlinear associations between two random variables.…”

“…Data screening steps were applied to select more reliable survival‐related genes using external datasets from ICGC via distance correlation. 24 , 25 Distance correlation measures linear and nonlinear associations between two random variables. The distance correlation between gene expression level and the linear predictor values of Cox regression models using clinical variables, such as patient age and tumor stage, were calculated.…”

Understanding How Cells Probe the World: A Preliminary Step towards Modeling Cell Behavior?

“…This strongly suggests that the efficiency of machine learning might be increased by suitable incorporation of biological knowledge. This is an incentive to try and understand the inner workings of modern data processing procedures that were rightly compared to blackboxes [67].…”