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

Hack, Joshua B.; Watkins, Joseph C.; Hammer, Michael F.

doi:10.1242/bio.060286

Cited by 3 publications

(1 citation statement)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… ABSTRACT First Person is a series of interviews with the first authors of a selection of papers published in Biology Open, helping researchers promote themselves alongside their papers. Joshua Hack is first author on ‘ Machine learning models reveal distinct disease subgroups and improve diagnostic and prognostic accuracy for individuals with pathogenic SCN8A gain-of-function variants ’, published in BiO. Joshua is a Data Scientist in the lab of Dr Michael Hammer at the University of Arizona, investigating the mechanisms of diseases and expanding the phenotypic landscapes of rare genetic diseases towards the goal of improving treatment strategies on an individual basis.…”

mentioning

confidence: 99%

First person – Joshua Hack

2024

Biology Open

View full text Add to dashboard Cite

First Person is a series of interviews with the first authors of a selection of papers published in Biology Open, helping researchers promote themselves alongside their papers. Joshua Hack is first author on ‘ Machine learning models reveal distinct disease subgroups and improve diagnostic and prognostic accuracy for individuals with pathogenic SCN8A gain-of-function variants’, published in BiO. Joshua is a Data Scientist in the lab of Dr Michael Hammer at the University of Arizona, investigating the mechanisms of diseases and expanding the phenotypic landscapes of rare genetic diseases towards the goal of improving treatment strategies on an individual basis.

show abstract

mentioning

confidence: 99%

First person – Joshua Hack

2024

Biology Open

View full text Add to dashboard Cite

show abstract

Machine Learning-Based Identification of High-Risk Patterns in Atrial Fibrillation Ablation Outcomes

Oloko-Oba,

Liu,

Wood

et al. 2024

Preprint

View full text Add to dashboard Cite

Background Atrial fibrillation (AF) is one of the most common types of cardiac arrhythmias, often leading to serious health issues such as stroke, heart failure, and higher mortality rates. Its global impact is rising due to aging populations and growing comorbidities, creating an urgent need for more effective treatment methods. AF ablation, a key treatment option, has success rates that vary widely among patients. Conventional predictors of ablation outcomes, which primarily rely on sociodemographic and clinical factors, fall short of capturing the heterogeneity within patient populations, highlighting the potential for data-driven methods to provide deeper insights into procedural success. Objectives To uncover meaningful patient subgroups based on AF ablation outcomes and identify diagnostic codes associated with failure. Methods Machine learning clustering with must-link and cannot-link constraints was applied to electronic health records to discover meaningful clusters, revealing patient-specific factors influencing procedural success or failure. Statistical analyses, including chi-square tests, were used to identify diagnostic codes significantly associated with ablation failure. Results Out of the 145 diagnostic codes examined, thirteen significant codes were identified and categorized into four primary risk groups, ranked by their impact on procedural outcomes: (1) direct contributors affecting cardiovascular health, (2) indirect factors that contribute to systemic stress, (3) complications related to anticoagulation and hemorrhagic risks that can impact bleeding management, and (4) broader health indicators reflecting a general health burden that reduce patients resilience to procedural stress. Conclusions This study shows the importance of cardiovascular and non-cardiovascular factors in AF ablation outcomes, emphasizing the need for a more comprehensive pre-procedural evaluation. It also contributes to the application of machine learning in personalized risk assessment for AF and advancing individualized care strategies that may improve ablation success.

show abstract

Patients carrying pathogenic SCN8A variants with loss‐ and gain‐of‐function effects can be classified into five subgroups exhibiting varying developmental and epileptic components of encephalopathy

Hack,

Watkins,

Schreiber

et al. 2024

Epilepsia

View full text Add to dashboard Cite

ObjectivePhenotypic heterogeneity presents challenges in providing clinical care to patients with pathogenic SCN8A variants, which underly a wide disease spectrum ranging from neurodevelopmental delays without seizures to a continuum of mild to severe developmental and epileptic encephalopathies (DEEs). An important unanswered question is whether there are clinically important subgroups within this wide spectrum. Using both supervised and unsupervised machine learning (ML) approaches, we previously found statistical support for two and three subgroups associated with loss‐ and gain‐ of‐ function vari‐ants, respectively. Here, we test the hypothesis that the unsupervised subgroups (U1–U3) are distinguished by differential contributions of developmental and epileptic components.MethodsWe predicted that patients in the U1 and U2 subgroups would differ in timing of developmental delay and seizure onset, with earlier and concurrent onset of both features for the U3 subgroup. Standard statistical procedures were used to test these predictions, as well as to investigate clinically relevant associations among all five subgroups.ResultsTwo‐population proportion and Kruskal–Wallis tests supported the hypothesis of a reversed order of developmental delay and seizure onset for patients in U1 and U2, and nearly synchronous developmental delay/seizure onset for the U3 (termed DEE) subgroup. Association testing identified subgroup variation in treatment response, frequency of initial seizure type, and comorbidities, as well as different median ages of developmental delay onset for all five subgroups.SignificanceUnsupervised ML approaches discern differential developmental and epileptic components among patients with SCN8A‐related epilepsy. Patients in U1 (termed developmental encephalopathy) typically gain seizure control yet rarely experience improvements in development, whereas those in U2 (termed epileptic encephalopathy) have fewer if any developmental impairments despite difficulty in achieving seizure control. This understanding improves prognosis and clinical management and provides a framework to discover mechanisms underlying variability in clinical outcome of patients with SCN8A‐related disorders.

show abstract

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

Cited by 3 publications

References 31 publications

First person – Joshua Hack

First person – Joshua Hack

Machine Learning-Based Identification of High-Risk Patterns in Atrial Fibrillation Ablation Outcomes

Patients carrying pathogenic SCN8A variants with loss‐ and gain‐of‐function effects can be classified into five subgroups exhibiting varying developmental and epileptic components of encephalopathy

Contact Info

Product

Resources

About