Detecting copy number variations (CNVs) in human genome using comparative datasets and machine learning models

Organisatsiooni nimi
Institute of Genomics, Research group of pharmacogenetics
Kokkuvõte
Copy number variations (CNVs) are deletions, duplications, inversions and insertions of DNA sequences in the genome. Despite their lower occurrence compared to other variation types in human genomes, such as single nucleotide polymorphisms (SNPs), they can result in different phenotypes with clinical significance in many cases. Accurate detection of CNVs is, therefore, an important and active research field. Due to the complex patterns and correlations in the genomic data, nonlinear data-driven methods are good candidates for modeling CNV signals. In preliminary analyses, we developed and trained models for this task with promising results using limited datasets. This project aims to build on these models to detect genome-wide and gene-specific CNVs potentially with state-of-the-art performance using genotype and whole genome sequencing data available in the Estonian Biobank.
Lõputöö kaitsmise aasta
2024-2025
Juhendaja
Burak Yelmen
Suhtlemiskeel(ed)
inglise keel
Nõuded kandideerijale
The candidate student should be comfortable programming with python and have at least a basic understanding of machine learning concepts, preferably with hands-on experience.
Tase
Magister
Märksõnad

Kandideerimise kontakt

 
Nimi
Burak Yelmen
Tel
E-mail
burak.yelmen@ut.ee