Predicting functional effects of ion channel variants using new phenotypic machine learning methods

Abstract

Missense variants in genes encoding ion channels are associated with a spectrum of severe diseases. Variant effects on biophysical function correlate with clinical features and can be categorized as gain- or loss-of-function. This information enables a timely diagnosis, facilitates precision therapy, and guides prognosis. Functional characterization presents a bottleneck in translational medicine. Machine learning models may be able to rapidly generate supporting evidence by predicting variant functional effects. Here, we describe a multi-task multi-kernel learning framework capable of harmonizing functional results and structural information with clinical phenotypes. This novel approach extends the human phenotype ontology towards kernel-based supervised machine learning. Our gain- or loss-of-function classifier achieves high performance (mean accuracy 0.853 SD 0.016, mean AU-ROC 0.912 SD 0.025), outperforming both conventional baseline and state-of-the-art methods. Performance is robust across different phenotypic similarity measures and largely insensitive to phenotypic noise or sparsity. Localized multi-kernel learning offered biological insight and interpretability by highlighting channels with implicit genotype-phenotype correlations or latent task similarity for downstream analysis.

Complex biophysical changes and reduced neuronal firing in anSCN8Avariant associated with developmental delay and epilepsy

“…Recently, clinical prediction tools were developed to facilitate the clinical interpretation of SCN variants (39)(40)(41). However, in agreement with the observed complex biophysical alterations, these tools were inconclusive regarding the functional ramification of SCN8A G1625R, predicting 58% LoF (39), 67% GoF (41) or GoF (40), highlighting the need for functional analysis.…”

“…The recurrent SCN8A missense c.4873G>A mutation, converting glycine 1625 to arginine in the S4 segment of the domain IV (Na V 1.6 G1625R ), has been independently identified in three patients with DEE [3,20,21]. This mutation was classified as pathogenic, predicted to cause GoF (67%) [22][23][24], but was not functionally characterized.…”

Machine learning models reveal distinct disease subgroups and improve diagnostic and prognostic accuracy for individuals with pathogenic SCN8A gain-of-function variants

“…Additional variants were included by utilizing the LOF Classifier ( Hack et al, 2023 ) to categorize variants: those with a probability of loss of function (LOF) [prob(LOF)] <0.3 were considered GOF, while those with a prob(LOF) >0.3 were considered intermediate or true LOF variants and excluded from the study. Alternative methods exist for classifying variants as GOF or LOF ( Bosselmann et al, 2023 ; Brunklaus et al, 2022 ; Heyne et al, 2020 ). After reviewing these alternatives, it was determined that the LOF classifier from Hack et al, 2023 was sufficient as justification for inclusion of variants in a clinically focused study.…”

Respiratory Dysfunction in Epileptic Encephalopathies: Insights and Challenges

“…Respiratory dysfunction, ranging from intermittent apneas to chronic hypoventilation, can significantly diminish the quality of life for individuals with epileptic encephalopathies. Sleep disturbances, daytime fatigue, and the need for ventilatory support devices can limit their ability to engage in typical daily activities and social interactions [ 32 ]. Chronic respiratory disturbances can exacerbate cognitive and behavioral challenges in epileptic encephalopathies.…”

“…The presence of comorbidities, such as cognitive impairment or cardiac abnormalities, can complicate the management of respiratory dysfunction and influence patient outcomes. Comprehensive care that addresses these comorbidities is essential [ 32 ].…”