Abstract:Understanding gene regulation is an important step to understanding how essential mechanisms are controlled in biological systems. One of the central goals of biology is to identify which transcription factors (TFs) regulate the transcription of which target genes and what are their downstream effects. Functional assays such as ChIP-seq and DNase I together can provide a TF binding map of TF binding sites on DNA. However, the binding alone may not result in changing the target gene expression. Thus, functional validation is necessary to show that the binding influences target gene expression. The standard approach to functional validation is to perform artificial TF knockdown experiments and declare the differentially expressed genes as validated target genes. Instead of artificial perturbation, we propose to leverage the naturally-occurring genetic variations as the source of perturbations that vary gene expressions and to analyze population SNP and geneexpression data in order to validate the TF binding map. Compared to the standard approach that perturbs TF concentration for a single TF at a time, our approach is potentially more powerful, because any aspects of the TF-target interaction, including TF concentration and TF binding affinity, can be perturbed by a large number of SNPs found across the genome. In addition, we are able to leverage existing SNP and gene expression data, which is available from the popular expression quantitative trait locus mapping studies. We introduce a statistical approach, based on conditional Gaussian Bayesian networks, that integrates population SNP and gene expression data with TF binding data to validate the TF binding map. We develop an efficient learning algorithm for learning the gene regulatory network by using the TF binding data as prior knowledge and selecting the TF-target interactions that are validated based on population SNP and gene-expression data. Given the estimated network, we perform inference on the estimated probabilistic graphical models to determine downstream genes that are affected by the TF-target interactions. We demonstrate our method on ENCODE ChIP-seq and DNase I data, and on population SNP and expression data from lymphoblastoid cells, originally collected for the 1000 Genomes and HapMap 3 projects respectively. Finally, we apply our approach to validate the TF binding map of ER and its coregulators in breast cancer using ENCODE ChIP-seq and DNase I data, and population SNP and expression data from the TCGA project.

A non-Gaussian factor analysis approach to transcription Network Component Analysis.

Bayesian non-negative factor analysis for reconstructing transcriptional regulatory network

Functional Validation of Transcription Factor to Gene Interactions by Statistical Learning of Gaussian Bayesian networks from SNP and Expression data .

An Iterated Conditional Modes solution for sparse Bayesian factor modeling of transcriptional regulatory networks

FISHFactor: A Probabilistic Factor Model for Spatial Transcriptomics Data with Subcellular Resolution

A Bayesian inference transcription factor activity model for the analysis of single-cell transcriptomes

Sufficient Direction Factor Model and Its Application to Gene Expression Quantitative Trait Loci Discovery

A sparse Bayesian factor model for the construction of gene co-expression networks from single-cell RNA sequencing count data

Semi-Confirmatory Factor Analysis for High-Dimensional Data with Interconnected Community Structures

An Efficient Nonnegative Matrix Factorization Model for Finding Cancer Associated Genes by Integrating Data from Genome, Transcriptome and Interactome.

Using Graphical Adaptive Lasso Approach to Construct Transcription Factor and Microrna'S Combinatorial Regulatory Network in Breast Cancer

Diffusion Component Analysis: Unraveling Functional Topology in Biological Networks

A novel constrained genetic algorithm-based Boolean network inference method from steady-state gene expression data

Composite Module Analyst: a fitness-based tool for identification of transcription factor binding site combinations

Sparse Nonnegative Matrix Factorization with Genetic Algorithms for Microarray Analysis

Neural Network-based Automatic Factor Construction

A Novel Fuzzy Bi-Clustering Algorithm with AFS for Identification of Co-Regulated Genes

A deep learning model to identify gene expression level using cobinding transcription factor signals

TIANA: transcription factors cooperativity inference analysis with neural attention

Identifiable and interpretable nonparametric factor analysis

Gaussian Process Modelling Of Latent Chemical Species: Applications To Inferring Transcription Factor Activities