Machine learning meets psoriasis: identifying key lactylation biomarkers as potential targets for diagnosis and therapies - PubMed
18 days ago
- #Machine Learning
- #Lactylation
- #Psoriasis Biomarkers
- Psoriasis research integrates machine learning to discover lactylation-related biomarkers, identifying MPHOSPH6, ENO1, MKI67, and FABP5 as key diagnostic targets validated by ROC curves and RT-qPCR.
- Analysis using WGCNA and differential gene screening pinpointed 26 lactylation-related genes from 1,623 psoriasis-associated genes, with immune infiltration showing significant correlations to immune cells in lesions.
- Mendelian randomization indicates that elevated levels of MPHOSPH6 and ENO1 act as risk factors for psoriasis, while the DSigDB database suggests 103 potential drugs targeting these biomarkers for therapeutic development.