Genetic analysis of ocular tumour-associated genes using large genomic datasets: insights into selection constraints and variant representation in the population

Abstract

BackgroundLarge genomic databases enable genetic evaluation in terms of haploinsufficiency and prevalence of missense and synonymous variants. We explored these parameters in ocular tumour-associated genes.MethodsA curated list of ocular tumour-associated genes was assessed using the genomic databases Genome Aggregation Database (gnomAD) and DatabasE of genomiC varIation and Phenotype in Humans using Ensembl Resources (DECIPHER) and compared with breast and lung cancer-associated gene lists. Haploinsufficiency was determined based on specific criteria: probability of loss of function index ≥0.9 in gnomAD, upper CI O/E limit <0.35 for loss of function variants in gnomAD and/or a DECIPHER pHaplo ≥0.86. UniProt was used for further gene characterisation, and gene ontology Protein Analysis THrough Evolutionary Relationships was explored for common biological pathways. In addition, we identified genes with under-representation/over-representation of missense/synonymous variants.ResultsFifty-seven genes were identified in association with ocular and extraocular tumours.Regarding haploinsufficiency, 41% of genes met the criteria for negative selection, with 57% categorised as tumour-suppressing and 39% as oncogenic. Most genes were involved in regulatory processes. Regarding triplosensitivity, 33% of genes reached significance and 83% of these were haploinsufficient. Analysis of variants revealed under-representation of missense variants in 23% of genes and over-representation of synonymous variants in 5% of genes. Ocular tumour-associated genes exhibited higher scores for haploinsufficiency and triplosensitivity compared with breast and lung cancer-associated genes. Pathway analysis revealed significant enrichment in cellular proliferation, differentiation and division. Encoded proteins of ocular tumour-associated genes were generally longer than the median of the UniProt database.ConclusionOur findings highlight the importance of negative selection in ocular tumour genes, supporting cranial gene conservation. This study provides insights into ocular tumourigenesis and future research avenues.

The benefit of a complete reference genome for cancer structural variant analysis

“…While there are significant improvements in variant calling when SVs are detected against the CHM13-T2T reference (Aganezov et al 2022), the benefits of using the complete genome reference have yet to be determined for somatic variants in a cancer context. Additionally, since CHM13-T2T is still being annotated, the annotated reference genomes GRCh37 and GRCh38 still hold value for ranking and characterizing SVs (Collins et al 2020; Tanner et al 2024). Using both a complete and annotated reference genome would allow researchers to identify novel oncogenic candidate variants across different cancer types.…”