Alterations in lipidome profiles distinguish early-onset hyperuricemia, gout, and the effect of urate-lowering treatment

Interpretable machine learning framework to predict gout associated with dietary fiber and triglyceride-glucose index

BACKGROUND

Gout prediction is essential for the development of individualized prevention and treatment plans. Our objective was to develop an efficient and interpretable machine learning (ML) model using the SHapley Additive exPlanation (SHAP) to link dietary fiber and triglyceride-glucose (TyG) index to predict gout.

METHODS

Using datasets from the National Health and Nutrition Examination Survey (NHANES) (2005-2018) population to study dietary fiber, the TyG index was used to predict gout. After evaluating the performance of six ML models and selecting the Light Gradient Boosting Machine (LGBM) as the optimal algorithm, we interpret the LGBM model for predicting gout using SHAP and reveal the decision-making process of the model.

RESULTS

An initial survey of 70,190 participants was conducted, and after a gradual exclusion process, 12,645 cases were finally included in the study. Selection of the best performing LGBM model for prediction of gout associated with dietary fiber and TyG index (Area under the ROC curve (AUC): 0.823, 95% confidence interval (CI): 0.798-0.848, Accuracy: 95.3%, Brier score: 0.077). The feature importance of SHAP values indicated that age was the most important feature affecting the model output, followed by uric acid (UA). The SHAP values showed that lower dietary fiber values had a more pronounced effect on the positive prediction of the model, while higher values of the TyG index had a more pronounced effect on the positive prediction of the model.

CONCLUSION

The interpretable LGBM model associated with dietary fiber and TyG index showed high accuracy, efficiency, and robustness in predicting gout. Increasing dietary fiber intake and lowering the TyG index are beneficial in reducing the potential risk of gout.

Examining the Association of Rare Allelic Variants in Urate Transporters SLC22A11, SLC22A13, and SLC17A1 with Hyperuricemia and Gout

Genetic variations in urate transporters play a significant role in determining human urate levels and have been implicated in developing hyperuricemia or gout. Polymorphism in the key urate transporters, such as ABCG2, URAT1, or GLUT9 was well-documented in the literature. Therefore in this study, our objective was to determine the frequency and effect of rare nonsynonymous allelic variants of SLC22A11, SLC22A13, and SLC17A1 on urate transport. In a cohort of 150 Czech patients with primary hyperuricemia and gout, we examined all coding regions and exon-intron boundaries of SLC22A11, SLC22A13, and SLC17A1 using PCR amplification and Sanger sequencing. For comparison, we used a control group consisting of 115 normouricemic subjects. To examine the effects of the rare allelic nonsynonymous variants on the expression, intracellular processing, and urate transporter protein function, we performed a functional characterization using the HEK293A cell line, immunoblotting, fluorescent microscopy, and site directed mutagenesis for preparing variants in vitro. Variants p.V202M (rs201209258), p.R343L (rs75933978), and p.P519L (rs144573306) were identified in the SLC22A11 gene (OAT4 transporter); variants p.R16H (rs72542450), and p.R102H (rs113229654) in the SLC22A13 gene (OAT10 transporter); and the p.W75C variant in the SLC17A1 gene (NPT1 transporter). All variants minimally affected protein levels and cytoplasmic/plasma membrane localization. The functional in vitro assay revealed that contrary to the native proteins, variants p.P519L in OAT4 (p ≤ 0.05), p.R16H in OAT10 (p ≤ 0.05), and p.W75C in the NPT1 transporter (p ≤ 0.01) significantly limited urate transport activity. Our findings contribute to a better understanding of (1) the risk of urate transporter-related hyperuricemia/gout and (2) uric acid handling in the kidneys.