Interpretable deep learning in single-cell omics DOI Creative Commons
Manoj M Wagle, Siqu Long, Carissa Chen

и другие.

Bioinformatics, Год журнала: 2024, Номер 40(6)

Опубликована: Июнь 1, 2024

Abstract Motivation Single-cell omics technologies have enabled the quantification of molecular profiles in individual cells at an unparalleled resolution. Deep learning, a rapidly evolving sub-field machine has instilled significant interest single-cell research due to its remarkable success analysing heterogeneous high-dimensional data. Nevertheless, inherent multi-layer nonlinear architecture deep learning models often makes them ‘black boxes’ as reasoning behind predictions is unknown and not transparent user. This stimulated increasing body for addressing lack interpretability models, especially data analyses, where identification understanding regulators are crucial interpreting model directing downstream experimental validations. Results In this work, we introduce basics concept interpretable learning. followed by review recent applied various research. Lastly, highlight current limitations discuss potential future directions.

Язык: Английский

The technological landscape and applications of single-cell multi-omics DOI Open Access
Alev Baysoy, Zhiliang Bai, Rahul Satija

и другие.

Nature Reviews Molecular Cell Biology, Год журнала: 2023, Номер 24(10), С. 695 - 713

Опубликована: Июнь 6, 2023

Язык: Английский

Процитировано

438
Gene regulatory network inference in the era of single-cell multi-omics DOI
Pau Badia-i-Mompel, Lorna Wessels, Sophia Müller‐Dott

и другие.

Nature Reviews Genetics, Год журнала: 2023, Номер 24(11), С. 739 - 754

Опубликована: Июнь 26, 2023

Язык: Английский

Процитировано

179
Network pharmacology: towards the artificial intelligence-based precision traditional Chinese medicine DOI Creative Commons
Peng Zhang, Dingfan Zhang, Wuai Zhou

и другие.

Briefings in Bioinformatics, Год журнала: 2023, Номер 25(1)

Опубликована: Ноя. 22, 2023

Abstract Network pharmacology (NP) provides a new methodological perspective for understanding traditional medicine from holistic perspective, giving rise to frontiers such as Chinese network (TCM-NP). With the development of artificial intelligence (AI) technology, it is key NP develop network-based AI methods reveal treatment mechanism complex diseases massive omics data. In this review, focusing on TCM-NP, we summarize involved into three categories: relationship mining, target positioning and navigating, present typical application TCM-NP in uncovering biological basis clinical value Cold/Hot syndromes. Collectively, our review researchers with an innovative overview progress its TCM perspective.

Язык: Английский

Процитировано

165
Single-cell biological network inference using a heterogeneous graph transformer DOI Creative Commons
Anjun Ma, Xiaoying Wang, Jingxian Li

и другие.

Nature Communications, Год журнала: 2023, Номер 14(1)

Опубликована: Фев. 21, 2023

Abstract Single-cell multi-omics (scMulti-omics) allows the quantification of multiple modalities simultaneously to capture intricacy complex molecular mechanisms and cellular heterogeneity. Existing tools cannot effectively infer active biological networks in diverse cell types response these external stimuli. Here we present DeepMAPS for network inference from scMulti-omics. It models scMulti-omics a heterogeneous graph learns relations among cells genes within both local global contexts robust manner using multi-head transformer. Benchmarking results indicate performs better than existing clustering construction. also showcases competitive capability deriving cell-type-specific lung tumor leukocyte CITE-seq data matched diffuse small lymphocytic lymphoma scRNA-seq scATAC-seq data. In addition, deploy webserver equipped with functionalities visualizations improve usability reproducibility analysis.

Язык: Английский

Процитировано

89
The diversification of methods for studying cell–cell interactions and communication DOI
Erick Armingol, Hratch Baghdassarian, Nathan E. Lewis

и другие.

Nature Reviews Genetics, Год журнала: 2024, Номер 25(6), С. 381 - 400

Опубликована: Янв. 18, 2024

Язык: Английский

Процитировано

51
Inferring gene regulatory network from single-cell transcriptomes with graph autoencoder model DOI Creative Commons
Jiacheng Wang, Yaojia Chen, Quan Zou

и другие.

PLoS Genetics, Год журнала: 2023, Номер 19(9), С. e1010942 - e1010942

Опубликована: Сен. 13, 2023

The gene regulatory structure of cells involves not only the relationship between two genes, but also cooperative associations multiple genes. However, most network inference methods for single cell focus on and infer relationships pairs ignoring global which is crucial to identify regulations in complex biological systems. Here, we proposed a graph-based Deep learning model Regulatory networks Inference among Genes (DeepRIG) from single-cell RNA-seq data. To learn structure, DeepRIG builds prior graph by transforming expression data into co-expression mode. Then it utilizes autoencoder embed information contained latent embeddings reconstruct network. Extensive benchmarking results demonstrate that can accurately outperform existing simulated real-cell networks. Additionally, applied samples human peripheral blood mononuclear triple-negative breast cancer, presented provide accurate cell-type-specific novel regulators progression inhibition.

Язык: Английский

Процитировано

42
A single-cell and spatial RNA-seq database for Alzheimer’s disease (ssREAD) DOI Creative Commons
Cankun Wang, Diana Acosta, Megan E. McNutt

и другие.

Nature Communications, Год журнала: 2024, Номер 15(1)

Опубликована: Июнь 6, 2024

Abstract Alzheimer’s Disease (AD) pathology has been increasingly explored through single-cell and single-nucleus RNA-sequencing (scRNA-seq & snRNA-seq) spatial transcriptomics (ST). However, the surge in data demands a comprehensive, user-friendly repository. Addressing this, we introduce RNA-seq database for disease (ssREAD). It offers broader spectrum of AD-related datasets, an optimized analytical pipeline, improved usability. The encompasses 1,053 samples (277 integrated datasets) from 67 scRNA-seq snRNA-seq studies, totaling 7,332,202 cells. Additionally, it archives 381 ST datasets 18 human mouse brain studies. Each dataset is annotated with details such as species, gender, region, disease/control status, age, AD Braak stages. ssREAD also provides analysis suite cell clustering, identification differentially expressed spatially variable genes, cell-type-specific marker genes regulons, spot deconvolution integrative analysis. freely available at https://bmblx.bmi.osumc.edu/ssread/ .

Язык: Английский

Процитировано

19
DeepKEGG: a multi-omics data integration framework with biological insights for cancer recurrence prediction and biomarker discovery DOI Creative Commons
Wei Lan,

Haibo Liao,

Qingfeng Chen

и другие.

Briefings in Bioinformatics, Год журнала: 2024, Номер 25(3)

Опубликована: Март 27, 2024

Abstract Deep learning-based multi-omics data integration methods have the capability to reveal mechanisms of cancer development, discover biomarkers and identify pathogenic targets. However, current ignore potential correlations between samples in integrating data. In addition, providing accurate biological explanations still poses significant challenges due complexity deep learning models. Therefore, there is an urgent need for a method explore provide model interpretability. Herein, we propose novel interpretable (DeepKEGG) recurrence prediction biomarker discovery. DeepKEGG, hierarchical module designed local connections neuron nodes interpretability based on relationship genes/miRNAs pathways. pathway self-attention constructed correlation different generate feature representation enhancing performance model. Lastly, attribution-based importance calculation utilized related interpretation Experimental results demonstrate that DeepKEGG outperforms other state-of-the-art 5-fold cross validation. Furthermore, case studies also indicate serves as effective tool The code available at https://github.com/lanbiolab/DeepKEGG.

Язык: Английский

Процитировано

17
MarsGT: Multi-omics analysis for rare population inference using single-cell graph transformer DOI Creative Commons
Xiaoying Wang,

Maoteng Duan,

Jingxian Li

и другие.

Nature Communications, Год журнала: 2024, Номер 15(1)

Опубликована: Янв. 6, 2024

Abstract Rare cell populations are key in neoplastic progression and therapeutic response, offering potential intervention targets. However, their computational identification analysis often lag behind major types. To fill this gap, we introduce MarsGT: Multi-omics Analysis for population inference using a Single-cell Graph Transformer. It identifies rare probability-based heterogeneous graph transformer on single-cell multi-omics data. MarsGT outperforms existing tools identifying cells across 550 simulated four real human datasets. In mouse retina data, it reveals unique subpopulations of bipolar Müller glia subpopulation. lymph node detects an intermediate B potentially acting as lymphoma precursors. melanoma MAIT-like impacted by high IFN-I response the mechanism immunotherapy. Hence, offers biological insights suggests strategies early detection disease.

Язык: Английский

Процитировано

16
Computational Methods for Single-Cell Multi-Omics Integration and Alignment DOI Creative Commons
Stefan Stanojevic, Yijun Li, Aleksandar Ristivojevic

и другие.

Genomics Proteomics & Bioinformatics, Год журнала: 2022, Номер 20(5), С. 836 - 849

Опубликована: Окт. 1, 2022

Abstract Recently developed technologies to generate single-cell genomic data have made a revolutionary impact in the field of biology. Multi-omics assays offer even greater opportunities understand cellular states and biological processes. The problem integrating different omics with very dimensionality statistical properties remains, however, quite challenging. A growing body computational tools is being for this task, leveraging ideas ranging from machine translation theory networks, represents another frontier on interface biology science. Our goal review provide comprehensive, up-to-date survey techniques integration multi-omics data, while making concepts behind each algorithm approachable non-expert audience.

Язык: Английский

Процитировано

41