Breast cancer risk SNPs and associated expression QTLs focusing Bangladeshi population: An in silico analysis

Background

Single Nucleotide Polymorphism (SNP)s in the genome and associated genes cause susceptibility to breast cancer, the most common cancer leading to death in women. Variation in different human races' genomes makes breast cancer prognosis challenging in terms of targeted drugs and therapies. The study aimed to compile the Breast cancer associated SNPs used for screening in the existing publications focusing Bangladeshi population, followed by the identification of Expression quantitative trait loci (eQTLs) associated with those SNPs from the eQTL database. eQTLs identify genes whose expression is regulated by specific SNPs. In silico characterization in terms of variant effect prediction, co-expression, Gene Ontology (GO) enrichment, protein-protein interaction, and sequence motif analysis narrowed down a specific set of candidate genes.

Methods

Published reports emphasizing the SNPs screened for Breast cancer in Bangladeshi population were analyzed in PancanQTL for identification of eQTLs which uses genotype and gene expression data from The Cancer Genome Atlas. The gene description and GO associated with identified eQTLs were retrieved from the Ensembl database and characterizations were performed using variant effect predictor, Coexpedia, MEME suite, and STRINGdb.

Results

It was found from the published reports that not all variants showed strong association with the disease in Bangladeshi population. The cis-eQTLs associated with reported SNPs tested so far on Bangladeshi population are ZNF575, MRPL42P5, C15orf57, C15orf62, NFATC3, XRCC1, C14orf153, CKB, BAG5, KLC1, MARK3. Among them only ZNF575 was enlisted as breast cancer associated eQTL and the rest are linked with other types of cancer. These genes are mostly associated in DNA-binding transcription factor activity, protein binding, Intracellular protein transport II, transferase activity. Protein-protein interaction could predict some functional partners to connect the eQTLs with respective SNP harboring genes. Taking the commonly screened genes for breast cancer as targets breast cancer associated cis and trans eQTLs along with the associated survival eQTLs have been retrieved from the database and a list of specific variants are recommended for future studies to get a more comprehensive scenario about the disease prognosis.

Conclusion

Since it was found from the existing literature that the commonly used variants are not always associated with all human races, this simple and precise in silico study was carried out to analyze publicly available data. This helped limit specific candidate genes and variants which will be helpful in future population-based screenings in understanding breast cancer prognosis with a provision to develop population specific personalized drug.