Inferring gene regulatory networks from single-cell gene expression data via deep multi-view contrastive learning

Task-augmented cross-view imputation network for partial multi-view incomplete multi-label classification

In real-world scenarios, multi-view multi-label learning often encounters the challenge of incomplete training data due to limitations in data collection and unreliable annotation processes. The absence of multi-view features impairs the comprehensive understanding of samples, omitting crucial details essential for classification. To address this issue, we present a task-augmented cross-view imputation network (TACVI-Net) for the purpose of handling partial multi-view incomplete multi-label classification. Specifically, we employ a two-stage network to derive highly task-relevant features to recover the missing views. In the first stage, we leverage the information bottleneck theory to obtain a discriminative representation of each view by extracting task-relevant information through a view-specific encoder-classifier architecture. In the second stage, an autoencoder based multi-view reconstruction network is utilized to extract high-level semantic representation of the augmented features and recover the missing data, thereby aiding the final classification task. Extensive experiments on five datasets demonstrate that our TACVI-Net outperforms other state-of-the-art methods.

MulNet: a scalable framework for reconstructing intra- and intercellular signaling networks from bulk and single-cell RNA-seq data

Gene expression involves complex interactions between DNA, RNA, proteins, and small molecules. However, most existing molecular networks are built on limited interaction types, resulting in a fragmented understanding of gene regulation. Here, we present MulNet, a framework that organizes diverse molecular interactions underlying gene expression data into a scalable multilayer network. Additionally, MulNet can accurately identify gene modules and key regulators within this network. When applied across diverse cancer datasets, MulNet outperformed state-of-the-art methods in identifying biologically relevant modules. MulNet analysis of RNA-seq data from colon cancer revealed numerous well-established cancer regulators and a promising new therapeutic target, miR-8485, along with several downstream pathways it governs to inhibit tumor growth. MulNet analysis of single-cell RNA-seq data from head and neck cancer revealed intricate communication networks between fibroblasts and malignant cells mediated by transcription factors and cytokines. Overall, MulNet enables high-resolution reconstruction of intra- and intercellular communication from both bulk and single-cell data. The MulNet code and application are available at https://github.com/free1234hm/MulNet.

GCLink: a graph contrastive link prediction framework for gene regulatory network inference

MOTIVATION

Gene regulatory networks (GRNs) unveil the intricate interactions among genes, pivotal in elucidating the complex biological processes within cells. The advent of single-cell RNA-sequencing (scRNA-seq) enables the inference of GRNs at single-cell resolution. However, the majority of current supervised network inference methods typically concentrate on predicting pairwise gene regulatory interaction, thus failing to fully exploit correlations among all genes and exhibiting limited generalization performance.

RESULTS

To address these issues, we propose a graph contrastive link prediction (GCLink) model to infer potential gene regulatory interactions from scRNA-seq data. Based on known gene regulatory interactions and scRNA-seq data, GCLink introduces a graph contrastive learning strategy to aggregate the feature and neighborhood information of genes to learn their representations. This approach reduces the dependence of our model on sample size and enhance its ability in predicting potential gene regulatory interactions. Extensive experiments on real scRNA-seq datasets demonstrate that GCLink outperforms other state-of-the-art methods in most cases. Furthermore, by pretraining GCLink on a source cell line with abundant known regulatory interactions and fine-tuning it on a target cell line with limited amount of known interactions, our GCLink model exhibits good performance in GRN inference, demonstrating its effectiveness in inferring GRNs from datasets with limited known interactions.

AVAILABILITY AND IMPLEMENTATION

The source code and data are available at https://github.com/Yoyiming/GCLink.

AttentionGRN: a functional and directed graph transformer for gene regulatory network reconstruction from scRNA-seq data

Single-cell RNA sequencing (scRNA-seq) enables the reconstruction of cell type-specific gene regulatory networks (GRNs), offering detailed insights into gene regulation at high resolution. While graph neural networks have become widely used for GRN inference, their message-passing mechanisms are often limited by issues such as over-smoothing and over-squashing, which hinder the preservation of essential network structure. To address these challenges, we propose a novel graph transformer-based model, AttentionGRN, which leverages soft encoding to enhance model expressiveness and improve the accuracy of GRN inference from scRNA-seq data. Furthermore, the GRN-oriented message aggregation strategies are designed to capture both the directed network structure information and functional information inherent in GRNs. Specifically, we design directed structure encoding to facilitate the learning of directed network topologies and employ functional gene sampling to capture key functional modules and global network structure. Our extensive experiments, conducted on 88 datasets across two distinct tasks, demonstrate that AttentionGRN consistently outperforms existing methods. Furthermore, AttentionGRN has been successfully applied to reconstruct cell type-specific GRNs for human mature hepatocytes, revealing novel hub genes and previously unidentified transcription factor-target gene regulatory associations.

DeepIMAGER: Deeply Analyzing Gene Regulatory Networks from scRNA-seq Data

Understanding the dynamics of gene regulatory networks (GRNs) across diverse cell types poses a challenge yet holds immense value in unraveling the molecular mechanisms governing cellular processes. Current computational methods, which rely solely on expression changes from bulk RNA-seq and/or scRNA-seq data, often result in high rates of false positives and low precision. Here, we introduce an advanced computational tool, DeepIMAGER, for inferring cell-specific GRNs through deep learning and data integration. DeepIMAGER employs a supervised approach that transforms the co-expression patterns of gene pairs into image-like representations and leverages transcription factor (TF) binding information for model training. It is trained using comprehensive datasets that encompass scRNA-seq profiles and ChIP-seq data, capturing TF-gene pair information across various cell types. Comprehensive validations on six cell lines show DeepIMAGER exhibits superior performance in ten popular GRN inference tools and has remarkable robustness against dropout-zero events. DeepIMAGER was applied to scRNA-seq datasets of multiple myeloma (MM) and detected potential GRNs for TFs of RORC, MITF, and FOXD2 in MM dendritic cells. This technical innovation, combined with its capability to accurately decode GRNs from scRNA-seq, establishes DeepIMAGER as a valuable tool for unraveling complex regulatory networks in various cell types.

SPREd: a simulation-supervised neural network tool for gene regulatory network reconstruction

Summary

Reconstruction of gene regulatory networks (GRNs) from expression data is a significant open problem. Common approaches train a machine learning (ML) model to predict a gene's expression using transcription factors' (TFs') expression as features and designate important features/TFs as regulators of the gene. Here, we present an entirely different paradigm, where GRN edges are directly predicted by the ML model. The new approach, named "SPREd," is a simulation-supervised neural network for GRN inference. Its inputs comprise expression relationships (e.g. correlation, mutual information) between the target gene and each TF and between pairs of TFs. The output includes binary labels indicating whether each TF regulates the target gene. We train the neural network model using synthetic expression data generated by a biophysics-inspired simulation model that incorporates linear as well as non-linear TF-gene relationships and diverse GRN configurations. We show SPREd to outperform state-of-the-art GRN reconstruction tools GENIE3, ENNET, PORTIA, and TIGRESS on synthetic datasets with high co-expression among TFs, similar to that seen in real data. A key advantage of the new approach is its robustness to relatively small numbers of conditions (columns) in the expression matrix, which is a common problem faced by existing methods. Finally, we evaluate SPREd on real data sets in yeast that represent gold-standard benchmarks of GRN reconstruction and show it to perform significantly better than or comparably to existing methods. In addition to its high accuracy and speed, SPREd marks a first step toward incorporating biophysics principles of gene regulation into ML-based approaches to GRN reconstruction.

Availability and implementation

Data and code are available from https://github.com/iiiime/SPREd.

GMFGRN: a matrix factorization and graph neural network approach for gene regulatory network inference

The recent advances of single-cell RNA sequencing (scRNA-seq) have enabled reliable profiling of gene expression at the single-cell level, providing opportunities for accurate inference of gene regulatory networks (GRNs) on scRNA-seq data. Most methods for inferring GRNs suffer from the inability to eliminate transitive interactions or necessitate expensive computational resources. To address these, we present a novel method, termed GMFGRN, for accurate graph neural network (GNN)-based GRN inference from scRNA-seq data. GMFGRN employs GNN for matrix factorization and learns representative embeddings for genes. For transcription factor-gene pairs, it utilizes the learned embeddings to determine whether they interact with each other. The extensive suite of benchmarking experiments encompassing eight static scRNA-seq datasets alongside several state-of-the-art methods demonstrated mean improvements of 1.9 and 2.5% over the runner-up in area under the receiver operating characteristic curve (AUROC) and area under the precision-recall curve (AUPRC). In addition, across four time-series datasets, maximum enhancements of 2.4 and 1.3% in AUROC and AUPRC were observed in comparison to the runner-up. Moreover, GMFGRN requires significantly less training time and memory consumption, with time and memory consumed <10% compared to the second-best method. These findings underscore the substantial potential of GMFGRN in the inference of GRNs. It is publicly available at https://github.com/Lishuoyy/GMFGRN.

SPREd: A simulation-supervised neural network tool for gene regulatory network reconstruction

Reconstruction of gene regulatory networks (GRNs) from expression data is a significant open problem. Common approaches train a machine learning (ML) model to predict a gene's expression using transcription factors' (TFs') expression as features and designate important features/TFs as regulators of the gene. Here, we present an entirely different paradigm, where GRN edges are directly predicted by the ML model. The new approach, named "SPREd" is a simulation-supervised neural network for GRN inference. Its inputs comprise expression relationships (e.g., correlation, mutual information) between the target gene and each TF and between pairs of TFs. The output includes binary labels indicating whether each TF regulates the target gene. We train the neural network model using synthetic expression data generated by a biophysics-inspired simulation model that incorporates linear as well as non-linear TF-gene relationships and diverse GRN configurations. We show SPREd to outperform state-of-the-art GRN reconstruction tools GENIE3, ENNET, PORTIA and TIGRESS on synthetic datasets with high co-expression among TFs, similar to that seen in real data. A key advantage of the new approach is its robustness to relatively small numbers of conditions (columns) in the expression matrix, which is a common problem faced by existing methods. Finally, we evaluate SPREd on real data sets in yeast that represent gold standard benchmarks of GRN reconstruction and show it to perform significantly better than or comparably to existing methods. In addition to its high accuracy and speed, SPREd marks a first step towards incorporating biophysics principles of gene regulation into ML-based approaches to GRN reconstruction.

Enhancing the prediction of IDC breast cancer staging from gene expression profiles using hybrid feature selection methods and deep learning architecture

Prediction of the stage of cancer plays an important role in planning the course of treatment and has been largely reliant on imaging tools which do not capture molecular events that cause cancer progression. Gene-expression data-based analyses are able to identify these events, allowing RNA-sequence and microarray cancer data to be used for cancer analyses. Breast cancer is the most common cancer worldwide, and is classified into four stages - stages 1, 2, 3, and 4 [2]. While machine learning models have previously been explored to perform stage classification with limited success, multi-class stage classification has not had significant progress. There is a need for improved multi-class classification models, such as by investigating deep learning models. Gene-expression-based cancer data is characterised by the small size of available datasets, class imbalance, and high dimensionality. Class balancing methods must be applied to the dataset. Since all the genes are not necessary for stage prediction, retaining only the necessary genes can improve classification accuracy. The breast cancer samples are to be classified into 4 classes of stages 1 to 4. Invasive ductal carcinoma breast cancer samples are obtained from The Cancer Genome Atlas (TCGA) and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) datasets and combined. Two class balancing techniques are explored, synthetic minority oversampling technique (SMOTE) and SMOTE followed by random undersampling. A hybrid feature selection pipeline is proposed, with three pipelines explored involving combinations of filter and embedded feature selection methods: Pipeline 1 - minimum-redundancy maximum-relevancy (mRMR) and correlation feature selection (CFS), Pipeline 2 - mRMR, mutual information (MI) and CFS, and Pipeline 3 - mRMR and support vector machine-recursive feature elimination (SVM-RFE). The classification is done using deep learning models, namely deep neural network, convolutional neural network, recurrent neural network, a modified deep neural network, and an AutoKeras generated model. Classification performance post class-balancing and various feature selection techniques show marked improvement over classification prior to feature selection. The best multiclass classification was found to be by a deep neural network post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with a Cohen-Kappa score of 0.303 and a classification accuracy of 53.1%. For binary classification into early and late-stage cancer, the best performance is obtained by a modified deep neural network (DNN) post SMOTE and random undersampling, and feature selection using mRMR and recursive feature elimination, with an accuracy of 81.0% and a Cohen-Kappa score (CKS) of 0.280. This pipeline also showed improved multiclass classification performance on neuroblastoma cancer data, with a best area under the receiver operating characteristic (auROC) curve score of 0.872, as compared to 0.71 obtained in previous work, an improvement of 22.81%. The results and analysis reveal that feature selection techniques play a vital role in gene-expression data-based classification, and the proposed hybrid feature selection pipeline improves classification performance. Multi-class classification is possible using deep learning models, though further improvement particularly in late-stage classification is necessary and should be explored further.

Transformer Architecture and Attention Mechanisms in Genome Data Analysis: A Comprehensive Review

The emergence and rapid development of deep learning, specifically transformer-based architectures and attention mechanisms, have had transformative implications across several domains, including bioinformatics and genome data analysis. The analogous nature of genome sequences to language texts has enabled the application of techniques that have exhibited success in fields ranging from natural language processing to genomic data. This review provides a comprehensive analysis of the most recent advancements in the application of transformer architectures and attention mechanisms to genome and transcriptome data. The focus of this review is on the critical evaluation of these techniques, discussing their advantages and limitations in the context of genome data analysis. With the swift pace of development in deep learning methodologies, it becomes vital to continually assess and reflect on the current standing and future direction of the research. Therefore, this review aims to serve as a timely resource for both seasoned researchers and newcomers, offering a panoramic view of the recent advancements and elucidating the state-of-the-art applications in the field. Furthermore, this review paper serves to highlight potential areas of future investigation by critically evaluating studies from 2019 to 2023, thereby acting as a stepping-stone for further research endeavors.