CRU 326 - Male Germ Cells: from Genes to Function - CP: Bioinformatic pipelines and integrated analyses

Basic data for this project

Description

Male infertility is a highly heterogeneous disease for which many causes are still unknown and are assumed to be of genetic origin. Identification of genetic variation that is associated with specific phenotypes significantly relies on the availability of large patient cohorts and access to their clinical and genetic data, all of which are present in the CRU. Over the last years, technological advances in sequencing techniques enabled to acquire increasing amounts of data. Novel methods comprise amongst others: whole exome and genome sequencing (WES/WGS), transcriptome sequencing on tissue or single cell scale as well as methylome studies. This not only increases demand for computational resources, but also requires in-depth bioinformatic expertise to adequately address each individual OMICs' challenges. While there are established workflows to analyse OMICs data such as bulk RNA-Seq and WES, there is a strong need to optimise bioinformatic pipelines. Specifically, analysing newly emerging single cell sequencing data is subject to intensive research. The latest methods are RNA velocity and latent time estimation. Due to their novelty, performing such analyses not only requires application of the algorithms but also validation and benchmarking of results to ensure robustness of conclusions drawn from such data. Furthermore, current methods for calling copy number variants (CNVs) in exomes are error-prone and show highly varying quality when applied to different data sets. We aim at developing a new algorithm to provide stable results using state-of-the-art machine learning techniques. With a centralised Core Project, we provide a consistently high standard of data processing and analysis. In the first funding period, more than 2900 samples for seven data types (CGH/SNP arrays, whole exome/genome sequencing, (sc)RNA sequencing, whole genome bisulfite sequencing) have been successfully analysed, ultimately fuelling into the Male Fertility Gene Atlas (MFGA), a comprehensive resource for researchers in the field of reproductive medicine. Consolidating all data in one project will enable integrated analyses of multiple OMICs and application of machine learning techniques to uncover previously unknown associations between biological markers and clinically relevant phenotypes.