Cancer Genetics and Epigenetics, 2025, Vol.13, No.1, 1-10 http://medscipublisher.com/index.php/cge 3 3 The Discovery of Low Penetrance Genes and GWAS 3.1 The contribution of genome-wide association studies (GWAS) to the discovery of CRC-related SNPS Genome-wide association studies (GWAS) have greatly changed people's understanding of the genetic mechanism of CRC by identifying many common low-penetrant single nucleotide polymorphisms (SNPS) associated with the risk of colorectal cancer (CRC) (Ma et al., 2013; Huyghe et al., 2018; Schmit et al., 2018). These studies were carried out in a large population and identified more than 100 independent related signals. This indicates that the characteristic of CRC being prone to onset is related to many genes, and also suggests that both common and rare genetic changes can affect the risk of disease (Huyghe et al., 2018; Schmit et al., 2018). The GWAS research method has also identified risk sites that were previously undetectable by the candidate gene research method, providing new insights into the biological processes involved in CRC, such as immune function, DNA repair, and intercellular signal transduction (Ma et al., 2013; Huyghe et al., 2018; Schmit et al., 2018). With the continuous increase of GWAS research data and the improvement of risk prediction models, it also provides a basis for formulating personalized screening programs through multi-gene risk scores (Schmit et al., 2018). 3.2 Common risk loci and their functional annotations There are some common risk sites that are closely related to colorectal cancer. It is particularly worth mentioning that the genetic changes at the position of 8q24.21 (rs6983267) are more common in patients with a family history of CRC, while the locus 8q23.3 (rs16892766) is associated with advanced tumors (Abuli et al., 2010). Other loci such as 16q22.2 (rs9929218) are associated with colorectal adenoma. In large-scale GWAS comprehensive analyses, loci such as 4q22.2, 5p15.33, 6p21.31 and 11q23 were also discovered (Abuli et al., 2010; Schmit et al., 2018). Studies on the functions of these loci have found that many risk-related genetic changes are in regions that do not encode proteins. They usually act like switches, affecting the activities of nearby genes related to cancer occurrence, DNA repair, and immune responses (Huyghe et al., 2018; Law et al., 2024). Recently, through fine localization and epigenetic studies, specific genetic changes have begun to be mapped to the genes they affect, making the molecular reasons for the high incidence of CRC clearer (Law et al., 2024). 3.3 Cumulative Impact of low-risk variations on individual disease susceptibility Although individual genetic changes with low penetrance only slightly increase the risk of colorectal cancer, because they are common in the population, the combined impact is significant (Houlston and Tomlinson, 2001; Ma et al., 2013). The polygenic risk score combines the effects of multiple common genetic changes to identify individuals with a significantly increased risk of CRC-based on genetic characteristics, up to 4.3% of the population has at least twice the possibility of developing CRC as others (Schmit et al., 2018). Genetic factor analysis shows that these low-risk genetic changes can explain a large part of the risk of familial CRC. Recent studies estimate that common genetic types can explain 14.7% of the difference in the risk of disease among family members (Ma et al., 2013; Schmit et al., 2018). With more and more genetic changes being discovered and incorporated into risk assessment models, the ability to predict the possibility of CRC onset has been continuously enhanced, which also makes personalized prevention and screening programs more achievable (Schmit et al., 2018; Huyghe et al., 2018). 4 Polygenic Risk Score (PRS) 4.1 Construction of the PRS weighted model based on multiple SNPS The polygenic risk Score (PRS) is derived by adding together the effects of many single nucleotide polymorphisms (SNPS) related to the risk of colorectal cancer (CRC), and each snp is assigned different weights according to the degree of influence obtained from genome-wide association studies (GWAS) (Thomas et al., 2020; Duenas et al., 2023; Tamlander et al., 2024). There are many methods for constructing PRS. For example, first select a group of verified CRC-related gene loci, and then use machine learning methods to handle the interrelationships between genes. There is also the Bayesian genome-wide risk prediction model (like LDpred), which can consider more than one million gene changes. Make risk estimation more accurate (Thomas et al., 2020; Duenas et al., 2023).
RkJQdWJsaXNoZXIy MjQ4ODYzNA==