Search bioRxivSearch

Biology subjects

Garcia, M.

Publications and source records attributed to Garcia, M..

2 recordsLinked to original sources

Sarek: A portable workflow for whole-genome sequencing analysis of germline and somatic variants

SummaryWhole-genome sequencing (WGS) is a cornerstone of precision medicine, but portable and reproducible open-source workflows for WGS analyses of germline and somatic variants are lacking. We present Sarek, a modular, comprehensive, and easy-to-install workflow, combining a range of software for the identification and annotation of single-nucleotide variants (SNVs), insertion and deletion variants (indels), structural variants, tumor sample heterogeneity, and karyotyping from germline or paired tumor/normal samples. Sarek is implemented in a bioinformatics workflow language (Nextflow) with Docker and Singularity compatible containers, ensuring easy deployment and full reproducibility at any Linux based compute cluster or cloud computing environment. Sarek supports the human reference genomes GRCh37 and GRCh38, and can readily be used both as a core production workflow at sequencing facilities and as a powerful stand-alone tool for individual research groups.\n\nAvailabilitySource code and instructions for local installation are available at GitHub (https://github.com/SciLifeLab/Sarek) under the MIT open-source license, and we invite the research community to contribute additional functionality as a collaborative open-source development project.

bioinformatics

Non-canonical aberrant DNA hypermethylation in glioma

Aberrant DNA hypermethylation is a hallmark of cancer although the underlying molecular mechanisms are still poorly understood. To study the possible role of 5-hydroxymethylcytosine (5hmC) in this process we analyzed the global and locus-specific genome-wide levels of 5hmC in primary samples from 54 gliomas and 72 colorectal cancer patients. Levels of 5hmC in colorectal cancer were very low and no consistent changes were detected between control tissues and tumors. As expected, levels of 5hmC in non-tumoral brain samples were high and significantly reduced at the 49,601 CpG sites in gliomas. Strikingly, hypo-hydroxymethylation at 4,627 (9.3%) of these CpG sites was associated with aberrant DNA hypermethylation. The DNA regions containing these CpG sites were enriched in H3K4me2, and presented a different genuine chromatin signature to that characteristic of the genes classically aberrantly hypermethylated in cancer. We conclude that this data identifies a novel 5hmC-dependent non-canonical class of aberrant DNA hypermethylation in glioma.

cancer biology