ÁNGEL ESTEBAN GIL¹, FERNANDO PÉREZ SANZ², JOSE GARCIA SOLANO³, Begoña Alburquerque González⁴, MARÍA ANTONIA PARREÑO GONZÁLEZ², MARÍA DEL CARMEN LEGAZ GARCÍA¹, Jesualdo Tomás Fernández Breis¹, Edith Rodríguez Braun³, Paola Patricia PIMENTEL CACERES³, Anne Tuomisto⁵, Markus Mäkinnen⁵, Ondrej Slaby⁶, Pablo CONESA ZAMORA³

(1) TECNOLOGÍAS DE MODELADO, PROCESAMIENTO Y GESTIÓN DEL CONOCIMIENTO, IMIB, España
(2) PLATAFORMA DE INFORMÁTICA BIOMÉDICA Y BIOINFORMÁTICA, IMIB, España
(3) PATOLOGÍA MOLECULAR Y FARMACOGENÉTICA, IMIB, España
(4) Universidad Católica San Antonio de Murcia (UCAM). Grupo IMIB: Patología Molecular y Farmacogenética, 30003, España (Región de Murcia)
(5) Department of Pathology. University of Oulu. Aapistie, 9. 90014 Oulu, Finlandia
(6) Central European Institute of Technology, Masaryk University / Department of Comprehensive Cancer Care, Masaryk Memorial Cancer Institute, Faculty of Medicine, Masaryk University, Kamenice 753/5, 625 00 Brno, República Checa

PMID:
Referencia: Esteban-Gil A, Pérez-Sanz F, García-Solano J, Alburquerque-González B, Parreño-González MA, Legaz-Garcia MC, Fernández-Breis, JT, Rodriguez-Braun E, Pimentel P, Tuomisto A, Mäkinen M, Slaby O, Conesa-Zamora P. 2019. ColPortal, an integrative multiomic platform for analysing epigenetic interactions in colorectal cancer. Scientific Data (Pending).
ISSN: 2052-4463
Revista: Scientific Data
Factor de impacto (2017): 9.8
Cuartil: 1

Colorectal cancer (CRC) is the third leading cause of cancer mortality worldwide. Different pathological pathways and molecular drivers have been described and some of the associated markers are used to select effective anti-neoplastic therapy. More recent evidence points to a causal role of microbiota and altered microRNA expression in CRC carcinogenesis, but their relationship with pathological drivers or molecular phenotypes is not clearly established. Joint analysis of clinical and omics data can help clarify such relations.

Our proposal is based on our previous results and experience working with biomedical data. ColPortal (https://colportal.imib.es) aims at providing an open web platform where users can download, visualize and analyse clinical and omic data in an integrated way. One of the main features of ColPortal is the possibility of making integrated analyses in real time. ColPortal has been developed using Java and R technologies and deployed in a HP Proliant DL360 G9 server with two Intel xeon processors with 12 cores (hyperthreading) and 256GB RAM. Currently, ColPortal allows the following types of analysis: Multiple Correspondence Analysis (MCA), Differential gene and miRNA expression analysis, Differential methylation analysis, Correlation analysis between microbiome and methylome.

Based on the principle that the histology is a complex manifestation of genetic and epigenetic changes, the main goal of the included series is to provide a representation of different histological subtypes of colorectal cancers with their associated "omic'' profiles. Those profiles have been obtained enriching certain subtypes, such as serrated adenocarcinoma and colorectal carcinomas, with histological features of microsatellite instability. Therefore, it does not represent the general frequency of these subtypes found amongst non-selected patients with colorectal carcinoma. The benefits of ColPortal for clinicians is the possibility of correlating not very common CRC subtypes with clinical features such as tumor presentation and prognosis; for pathologists it allows to associate histological characteristic of CRC with particular molecular changes. Finally, researchers can generate proof of concepts and establish relationships amongst genetics and epigenetics features at different levels (e.g. microbiota and microtranscriptomics; microbiota and methylomics, etc).

We present ColPortal, a platform that integrates transcriptomic, microtranscriptomic, methylomic and microbiota data of patients with colorectal cancer. ColPortal also includes detailed information of histological features and digital histological slides from the study cases, since histology is a morphological manifestation of a complex molecular change. The current cohort consists of Caucasian patients from Europe. For each patient, demographic information, location, histology, tumor staging, tissue prognostic factors, molecular biomarker status and clinical outcomes are integrated with omics data. ColPortal allows one to perform multiomics analyses for groups of patients selected by their clinical data.