Independent Component Analysis (ICA) based-clustering of temporal RNA-seq data

Nascimento, Moysés; Silva, Fabyano Fonseca e; Sáfadi, Thelma; Nascimento, Ana Carolina Campana; Ferreira, Talles Eduardo Maciel; Barroso, Laís Mayara Azevedo; Azevedo, Camila Ferreira; Guimarães, Simone Eliza Faccione; Serão, Nick V. L.

doi:10.1371/journal.pone.0181195

Cited by 32 publications

(20 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SC-ION builds on our MATLAB-based pipeline Regression Tree Pipeline for Spatial, Temporal, and Replicate data (RTP-STAR) (7,65), which is an adaptation of the GENIE3 (20) network inference method and functions only on transcriptomic datasets. In SC-ION, we further improve on RTP-STAR by: 1) incorporating Dynamic Time Warping (DTW) clustering for temporal data (13) and Independent Component Analysis (ICA) clustering for non-temporal data (37); 2) allowing the user to provide separate regulator and target matrices for integration of DE gene-products; 3) integrating any number of different types of expression profiles into one GRN; and 4) providing our pipeline as an Abundance GRN Phosphosite GRN RShiny GUI ( Fig. 2A).…”

Section: Resultsmentioning

confidence: 99%

“…TF-centered gene regulatory networks were inferred using SC-ION version 2.0 (https://github.com/nmclark2/SCION). SC-ION builds on the RTP-STAR pipeline (7,65) by incorporating DTW and ICA clustering (13,37) and integration of different data types. SC-ION uses an adapted version of GENIE3 (20) which allows for separate regulator and target data matrices (54).…”

Section: R a F Tmentioning

confidence: 99%

See 1 more Smart Citation

Integrated omics networks reveal the temporal signaling events of brassinosteroid response inArabidopsis

Clark

Nolan

Wang

et al. 2020

Preprint

View full text Add to dashboard Cite

Brassinosteroids (BRs) are plant steroid hormones that are known to regulate cell division and stress response. We used a systems biology approach to integrate multi-omic datasets and unravel the molecular signaling events of BR response in Arabidopsis. We profiled the levels of 32,549 transcripts, 9,035 protein groups, and 26,950 phosphorylation sites from Arabidopsis seedlings treated with brassinolide (BL, most active BR) for six different lengths of time. We then constructed a network inference pipeline called Spatiotemporal Clustering and Inference of Omics Networks (SC-ION) that was able to integrate these multi-omic data into one, cohesive regulatory network. Our network illustrates the signaling cascade of BR response, starting with kinase-substrate phosphorylation and ending with transcriptional regulation. We used our network predictions to identify putative, relevant phosphorylation sites on the TF BRI1-EMSSUPPRESSOR 1 (BES1); the importance of which we experimentally validated. Additionally, we identified an uncharacterized TF, which we named BRONTOSAURUS (BRON), that regulates cell division. Further, we show that bron mutant roots are hypersensitive to BL. Together, this work demonstrates the power of integrative network analysis applied to multiomic data and provides fundamental insights into the molecular signaling events occurring during BR response.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: R a F Tmentioning

confidence: 99%

Integrated omics networks reveal the temporal signaling events of brassinosteroid response inArabidopsis

Clark

Nolan

Wang

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In general, the intent of the independent component analysis (ICA) is to find the hidden ‘independent component’ that refers to the gene module in this research [ 26 ]. When applied in the field of gene module detection, ICA usually splits express data matrix X into two matrices: a source matrix S and a mixing matrix A , which means is shown in Fig.…”

Section: Modularboost Methodsmentioning

confidence: 99%

ModularBoost: an efficient network inference algorithm based on module decomposition

Zhang

et al. 2021

BMC Bioinformatics

View full text Add to dashboard Cite

Background Given expression data, gene regulatory network(GRN) inference approaches try to determine regulatory relations. However, current inference methods ignore the inherent topological characters of GRN to some extent, leading to structures that lack clear biological explanation. To increase the biophysical meanings of inferred networks, this study performed data-driven module detection before network inference. Gene modules were identified by decomposition-based methods. Results ICA-decomposition based module detection methods have been used to detect functional modules directly from transcriptomic data. Experiments about time-series expression, curated and scRNA-seq datasets suggested that the advantages of the proposed ModularBoost method over established methods, especially in the efficiency and accuracy. For scRNA-seq datasets, the ModularBoost method outperformed other candidate inference algorithms. Conclusions As a complicated task, GRN inference can be decomposed into several tasks of reduced complexity. Using identified gene modules as topological constraints, the initial inference problem can be accomplished by inferring intra-modular and inter-modular interactions respectively. Experimental outcomes suggest that the proposed ModularBoost method can improve the accuracy and efficiency of inference algorithms by introducing topological constraints.

show abstract

“…Recently ICA was used to engineer features for further use in cancer-related classification tasks, using naïve Bayes classifier [58]. In Reference [59], ICA was used as a data pre-processing step in order to improve the clustering of temporal RNA-Seq data. It was suggested to use ICA in combination with wavelet-based data transformation in order to engineer transcriptomic features at “multiple resolution” [60] and use them to improve tumor classification and biomarker discovery.…”

Section: Applications Of Ica In Cancer Researchmentioning

confidence: 99%

Independent Component Analysis for Unraveling the Complexity of Cancer Omics Datasets

Sompairac

Nazarov

Czerwińska

et al. 2019

IJMS

View full text Add to dashboard Cite

Independent component analysis (ICA) is a matrix factorization approach where the signals captured by each individual matrix factors are optimized to become as mutually independent as possible. Initially suggested for solving source blind separation problems in various fields, ICA was shown to be successful in analyzing functional magnetic resonance imaging (fMRI) and other types of biomedical data. In the last twenty years, ICA became a part of the standard machine learning toolbox, together with other matrix factorization methods such as principal component analysis (PCA) and non-negative matrix factorization (NMF). Here, we review a number of recent works where ICA was shown to be a useful tool for unraveling the complexity of cancer biology from the analysis of different types of omics data, mainly collected for tumoral samples. Such works highlight the use of ICA in dimensionality reduction, deconvolution, data pre-processing, meta-analysis, and others applied to different data types (transcriptome, methylome, proteome, single-cell data). We particularly focus on the technical aspects of ICA application in omics studies such as using different protocols, determining the optimal number of components, assessing and improving reproducibility of the ICA results, and comparison with other popular matrix factorization techniques. We discuss the emerging ICA applications to the integrative analysis of multi-level omics datasets and introduce a conceptual view on ICA as a tool for defining functional subsystems of a complex biological system and their interactions under various conditions. Our review is accompanied by a Jupyter notebook which illustrates the discussed concepts and provides a practical tool for applying ICA to the analysis of cancer omics datasets.

show abstract

Independent Component Analysis (ICA) based-clustering of temporal RNA-seq data

Cited by 32 publications

References 26 publications

Integrated omics networks reveal the temporal signaling events of brassinosteroid response inArabidopsis

Integrated omics networks reveal the temporal signaling events of brassinosteroid response inArabidopsis

ModularBoost: an efficient network inference algorithm based on module decomposition

Independent Component Analysis for Unraveling the Complexity of Cancer Omics Datasets

Contact Info

Product

Resources

About