Artificial Reasoning Approaches for Predicting Hemophilia a Severity

Background: Hemophilia A (HA) is an X-linked congenital bleeding disorder, which leads to deficiency of clotting factor VIII (FVIII). Even though it mostly affects males, and females are considered carriers, variants in F8 in females can result in HA. Many females go undiagnosed and untreated for HA and their bleeding complications are attributed to other causes. Predicting the severity of HA for female patients can provide valuable insights for treating conditions associated with the disease such as heavy bleeding.

Objective: To predict the severity of HA in women based on F8 genotype and identify genotypes with a higher association for severe disease using artificial reasoning (AR) approaches.

Methods: Using multiple datasets of variants in the F8 and disease severity various repositories (e.g., My Life Our Future (MLOF)), we derived the sequence for the Factor VIII (FVIII) protein. Using the derived sequences, we used causal learning and artificial reasoning approaches such as ML classification models and Explainable AI (XAI) to predict the severity of HA in female patients and highlight genotypes with a higher association for severe HA.

Results: We apply causal learning to observational data to highlight the specific genotypes that have a direct cause-and-effect relation with severe HA in women. Utilizing different machine learning (ML) classification models, we validated our approach with predictive F1-scores of 0.88, 0.99, 0.93, 0.99 and 0.90 for all the validation sets. We picked the top-performing ML model to apply XAI via SHAP to identify genotypes with a higher association with severe HA.

Conclusion: AR-based approaches were successfully employed to predict HA severity in females based on variants in the F8. This study confirms previous research findings that ML can help predict the severity of hemophilia and provides new insights into genotypes with a higher association for severe disease. These results can be valuable for future studies in female patients with HA which is an urgent unmet need.

No relevant conflicts of interest to declare.