Gene regulation

Kipoi: accelerating the community exchange and reuse of predictive models for genomics

This entry was posted in deep_learning Gene regulation on by

Our next meeting will be at 1pm on Oct 1st, in room 4130 of the Discovery building. Our Selected paper is Kipoi: accelerating the community exchange and reuse of predictive models for genomics.
The abstract is as follows.

Advanced machine learning models applied to large-scale genomics datasets hold the promise to be major drivers for genome science. Once trained, such models can serve as a tool to probe the relationships between data modalities, including the effect of genetic variants on phenotype. However, lack of standardization and limited accessibility of trained models have hampered their impact in practice. To address this, we present Kipoi, a collaborative initiative to define standards and to foster reuse of trained models in genomics. Already, the Kipoi repository contains over 2,000 trained models that cover canonical prediction tasks in transcriptional and post-transcriptional gene regulation. The Kipoi model standard grants automated software installation and provides unified interfaces to apply and interpret models. We illustrate Kipoi through canonical use cases, including model benchmarking, transfer learning, variant effect prediction, and building new models from existing ones. By providing a unified framework to archive, share, access, use, and build on models developed by the community, Kipoi will foster the dissemination and use of machine learning models in genomics.

We welcome all who can join us for this discussion. Feel free to begin that discussion in the comments section below.

Virtual ChIP-seq: predicting transcription factor binding by learning from the transcriptome

This entry was posted in Gene regulation network_inference TF_activity on by

Our next meeting will be at 2pm on Mar 12th, in room 4160 of the Discovery building. Our Selected paper is Virtual ChIP-seq: predicting transcription factor binding by learning from the transcriptome.
.
The abstract is as follows.

Motivation: Identifying transcription factor binding sites is the first step in pinpointing non-coding mutations that disrupt the regulatory function of transcription factors and promote disease. ChIP-seq is the most common method for identifying binding sites, but performing it on patient samples is hampered by the amount of available biological material and the cost of the experiment. Existing methods for computational prediction of regulatory elements primarily predict binding in genomic regions with sequence similarity to known transcription factor sequence preferences. This has limited efficacy since most binding sites do not resemble known transcription factor sequence motifs, and many transcription factors are not even sequence-specific.

Results: We developed Virtual ChIP-seq, which predicts binding of individual transcription factors in new cell types using an artificial neural network that integrates ChIP-seq results from other cell types and chromatin accessibility data in the new cell type. Virtual ChIP-seq also uses learned associations between gene expression and transcription factor binding at specific genomic regions. This approach outperforms methods that use transcription factor sequence preferences in the form of position weight matrices, predicting binding for transcription factors (accuracy > 0.99; Matthews correlation coefficient > 0.3). In at least one validation cell type, performance of Virtual ChIP-seq is higher than all participants of the DREAM Challenge for in vivo transcription factor binding site prediction in 4 of 9 transcription factors that we could compare to.

 

We welcome all who can join us for this discussion. Feel free to begin that discussion in the comments section below.

Clustering gene expression time series data using an infinite Gaussian process mixture model

This entry was posted in clustering Gene regulation Time Series on by

Our next meeting will be at 2pm on Feb 12th, in room 4160 of the Discovery building. Our Selected paper is Clustering gene expression time series data using an infinite Gaussian process mixture model.
.
The abstract is as follows.

Transcriptome-wide time series expression profiling is used to characterize the cellular response to environmental perturbations. The first step to analyzing transcriptional response data is often to cluster genes with similar responses. Here, we present a nonparametric model-based method, Dirichlet process Gaussian process mixture model (DPGP), which jointly models data clusters with a Dirichlet process and temporal dependencies with Gaussian processes. We demonstrate the accuracy of DPGP in comparison to state-of-the-art approaches using hundreds of simulated data sets. To further test our method, we apply DPGP to published microarray data from a microbial model organism exposed to stress and to novel RNA-seq data from a human cell line exposed to the glucocorticoid dexamethasone. We validate our clusters by examining local transcription factor binding and histone modifications. Our results demonstrate that jointly modeling cluster number and temporal dependencies can reveal shared regulatory mechanisms. DPGP software is freely available online at https://github.com/PrincetonUniversity/DP_GP_cluster.

We welcome all who can join us for this discussion. Feel free to begin that discussion in the comments section below.

Learning causal networks with latent variables from multivariate information in genomic data

This entry was posted in causal_networks Gene regulation on by

Our next meeting will be at 11:00 on Dec 5th, in room 4160 of the Discovery building. Our Selected paper is Learning causal networks with latent variables from multivariate information in genomic data.
The abstract is as follows.

Learning causal networks from large-scale genomic data remains challenging in absence of time series or controlled perturbation experiments. We report an information- theoretic method which learns a large class of causal or non-causal graphical models from purely observational data, while including the effects of unobserved latent variables, commonly found in many genomic datasets. Starting from a complete graph, the method iteratively removes dispensable edges, by uncovering significant information contributions from indirect paths, and assesses edge-specific confidences from randomization of available data. The remaining edges are then oriented based on the signature of causality in observational data. The approach and associated algorithm, miic, outperform earlier methods on a broad range of benchmark networks. Causal network reconstructions are presented at different biological size and time scales, from gene regulation in single cells to whole genome duplication in tumor development as well as long term evolution of vertebrates. Miic is publicly available at https://github.com/miicTeam/MIIC.

We welcome all who can join us for this discussion. Feel free to begin that discussion in the comments section below.

Knowledge-guided gene prioritization reveals new insights into the mechanisms of chemoresistance

This entry was posted in Gene regulation network_inference on by

Our next meeting will be at 11:00 on Oct 10th, in room 4160 of the Discovery building. Our Selected paper is Knowledge-guided gene prioritization reveals new insights into the mechanisms of chemoresistance.
The abstract is as follows.

Background: Identification of genes whose basal mRNA expression predicts the sensitivity of tumor cells to cytotoxic treatments can play an important role in individualized cancer medicine. It enables detailed characterization of the mechanism of action of drugs. Furthermore, screening the expression of these genes in the tumor tissue may suggest the best course of chemotherapy or a combination of drugs to overcome drug resistance.

Results: We developed a computational method called ProGENI to identify genes most associated with the variation of drug response across different individuals, based on gene expression data. In contrast to existing methods, ProGENI also utilizes prior knowledge of protein–protein and genetic interactions, using random walk techniques. Analysis of two relatively new and large datasets including gene expression data on hundreds of cell lines and their cytotoxic responses to a large compendium of drugs reveals a significant improvement in prediction of drug sensitivity using genes identified by ProGENI compared to other methods. Our siRNA knockdown experiments on ProGENI-identified genes confirmed the role of many new genes in sensitivity to three chemotherapy drugs: cisplatin, docetaxel, and doxorubicin. Based on such experiments and extensive literature survey, we demonstrate that about 73% of our top predicted genes modulate drug response in selected cancer cell lines. In addition, global analysis of genes associated with groups of drugs uncovered pathways of cytotoxic response shared by each group.

Conclusions: Our results suggest that knowledge-guided prioritization of genes using ProGENI gives new insight into mechanisms of drug resistance and identifies genes that may be targeted to overcome this phenomenon.

We welcome all who can join us for this discussion. Feel free to begin that discussion in the comments section below.

Genome-scale high-resolution mapping of activating and repressive nucleotides in regulatory regions

This entry was posted in Abstracts Epigenome Gene regulation on by

Our next meeting will be at 3:00 on December 5th, in room 3160 of the Discovery building. Our Selected paper is Genome-scale high-resolution mapping of activating and repressive nucleotides in regulatory regions.
The abstract is as follows.

Massively parallel reporter assays (MPRAs) enable nucleotide-resolution dissection of transcriptional regulatory regions, such as enhancers, but only few regions at a time. Here we present a combined experimental and computational approach, Systematic high-resolution activation and repression profiling with reporter tiling using MPRA (Sharpr-MPRA), that allows high-resolution analysis of thousands of regions simultaneously. Sharpr-MPRA combines dense tiling of overlapping MPRA constructs with a probabilistic graphical model to recognize functional regulatory nucleotides, and to distinguish activating and repressive nucleotides, using their inferred contribution to reporter gene expression. We used Sharpr-MPRA to test 4.6 million nucleotides spanning 15,000 putative regulatory regions tiled at 5-nucleotide resolution in two human cell types. Our results recovered known cell-type-specific regulatory motifs and evolutionarily conserved nucleotides, and distinguished known activating and repressive motifs. Our results also showed that endogenous chromatin state and DNA accessibility are both predictive of regulatory function in reporter assays, identified retroviral elements with activating roles, and uncovered ‘attenuator’ motifs with repressive roles in active chromatin.