POPTric: Pathway-based Order Preserving Triclustering for gene sample time data analysis

• Published in: Expert systems with applications 2022-04, Vol.192, p.116336, Article 116336
• Main Authors: Mandal, Koyel, Sarmah, Rosy, Bhattacharyya, Dhruba Kumar
• Format: Article
• Language: English
• Subjects: Algorithms; Bioinformatics; Breast cancer; Breast cancer data; Data analysis; Data mining; Datasets; Enrichment analysis; Gene expression; Gene sample time data; Hub gene identification; Pathway; Regulatory mechanisms (biology); Semi-supervised triclustering
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2021.116336

Exploratory analysis of high throughput gene sample time (GST) data has an impact in biomedical and bioinformatics research. Mining gene expression pattern in such three dimensional data facilitate in understanding hidden biological knowledge as well as underlying complex gene regulatory mechanism. In particular, we propose a novel semi-supervised Pathway-based Order Preserving Triclustering (POPTric) algorithm which records gene expression pattern under subset of sample across subset of time points. Nowadays, integration of external knowledge has gaining popularity in the field of gene expression data analysis. Our algorithm is able to identify different types of triclusters such as additive, multiplicative, and additive-and-multiplicative triclusters. POPTric at first identifies biclusters and then it discovers triclusters from the 3D data. We apply our algorithm in artificial datasets as well as breast cancer, HIV datasets, and compare with other state-of-the-art methods. To establish the biological significance of proposed algorithm we perform Gene Ontology enrichment analysis. We also identify hub genes from the tricluster which are previously known to be associated with breast cancer. •A semi-supervised triclustering algorithm which is guided by KEGG pathway is presented here.•To evaluate the effectiveness of the proposed algorithm synthetic dataset is used.•The case study used breast cancer and HIV three dimensional gene expression data.•The proposed algorithm shows better performance than state-of-the-art methods.