首页|Research from Cardiff University in the Area of Machine Learning Described (Util ising biological experimental data and molecular dynamics for the classification of mutational hotspots through machine learning)
Research from Cardiff University in the Area of Machine Learning Described (Util ising biological experimental data and molecular dynamics for the classification of mutational hotspots through machine learning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News – Research findings on artificial intelligence are discussed in a new report. According to news reporting originating from Cardiff, United Kingdom, by NewsRx correspondents, research stated, “Motivation: Benzo [a]pyrene, a notorious DNA-damaging carcinogen, belongs to the family of polycyclic aromatic hydrocarbons commonly found in tobacco smoke. Sur prisingly, nucleotide excision repair (NER) machinery exhibits inefficiency in r ecognizing specific bulky DNA adducts including Benzo[a] pyrene Diol-Epoxide (BPDE), a Benzo[a]pyre ne metabolite.” The news journalists obtained a quote from the research from Cardiff University: “While sequence context is emerging as the leading factor linking the inadequat e NER response to BPDE adducts, the precise structural attributes governing thes e disparities remain inadequately understood. We therefore combined the domains of molecular dynamics and machine learning to conduct a comprehensive assessment of helical distortion caused by BPDE-Guanine adducts in multiple gene contexts. Specifically, we implemented a dual approach involving a random forest classifi cation-based analysis and subsequent feature selection to identify precise topol ogical features that may distinguish adduct sites of variable repair capacity. O ur models were trained using helical data extracted from duplexes representing b oth BPDE hotspot and non-hotspot sites within the TP53 gene, then applied to sit es within TP53, cII, and lacZ genes. We show our optimised model consistently ac hieved exceptional performance, with accuracy, precision, and f1 scores exceedin g 91 %. Our feature selection approach uncovered that discernible va riance in regional base pair rotation played a pivotal role in informing the dec isions of our model. Notably, these disparities were highly conserved among TP53 and lacZ duplexes and appeared to be influenced by the regional GC content.”
Cardiff UniversityCardiffUnited King domEuropeCyborgsEmerging TechnologiesMachine LearningMolecular Dynamic sPhysics
NETL
NSTL
