MLb-LDLr

• Published in: JACC. Basic to translational science 2021-11, Vol.6 (11), p.815-827
• Main Authors: Larrea-Sebal, Asier, Benito-Vicente, Asier, Fernandez-Higuero, José A., Jebari-Benslaiman, Shifa, Galicia-Garcia, Unai, Uribe, Kepa B., Cenarro, Ana, Ostolaza, Helena, Civeira, Fernando, Arrasate, Sonia, González-Díaz, Humberto, Martín, César
• Format: Article
• Language: English
• Subjects: familial hypercholesterolemia; LDL receptor; machine learning software; pathogenicity; prediction
• ISSN: 2452-302X; 2452-302X
• DOI: 10.1016/j.jacbts.2021.08.009

[Display omitted] •A machine-learning model has been developed to improve accuracy on predicting the activity of missense LDLr mutations.•ClinVar was used as database, and the model function was defined by using specific characteristics of the LDLr.•A high-score prediction ML model with specificity of 92.5% and sensitivity of 91.6% has been developed to predict pathogenicity of LDLr variants.•Implementation of high-predicting capacity software constitutes a valuable approach for assessing pathogenicity of LDLr variants to help in the early diagnosis and management of FH disease.•An open-access predictive software (MLb-LDLr) is provided to the scientific community. Untreated familial hypercholesterolemia (FH) leads to atherosclerosis and early cardiovascular disease. Mutations in the low-density lipoprotein receptor (LDLr) gene constitute the major cause of FH, and the high number of mutations already described in the LDLr makes necessary cascade screening or in vitro functional characterization to provide a definitive diagnosis. Implementation of high-predicting capacity software constitutes a valuable approach for assessing pathogenicity of LDLr variants to help in the early diagnosis and management of FH disease. This work provides a reliable machine learning model to accurately predict the pathogenicity of LDLr missense variants with specificity of 92.5% and sensitivity of 91.6%.