Skip to main content
Fig. 3 | BioData Mining

Fig. 3

From: Machine learning approaches for the genomic prediction of rheumatoid arthritis and systemic lupus erythematosus

Fig. 3

SHAP summary graph of top 20 SNPs of machine learning models. A) gradient tree boosting and (B) extreme gradient boosting models. As the SHAP value of an SNP (x-axis) increased, the probability of RA increased; the lower is the x-axis SHAP value of an SNP, the higher is the probability of SLE development. Each dot on the SHAP plot was calculated using the prediction model for each SNP’s attribution value for a participant. Dots are illustrated according to the feature values of each participant and accumulate vertically to indicate density. Blue represents 0/0 and red represents 0/1 or 1/1. SNP: single nucleotide polymorphism; SHAP: SHapley Additive exPlanation

Back to article page