Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

CMSV: Long-Read-Based Structural Variation Detection Through a CNN-Mamba Model.

Song Cheng¹, Hongbing Ma^1,2

¹School of Computer Science and Technology, Xinjiang University, Urumqi 830046, China.

|June 26, 2026

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Autoantibodies combined with systemic inflammation markers for predicting bone metastases in non-small cell lung cancer patients.

Frontiers in immunology·2026

Same author

Evolving computational paradigms for noncoding variant pathogenicity prediction.

Frontiers in molecular biosciences·2026

Same author

Elucidating the Gas Phase Thermochemistry and the H‑Atom Abstraction Reactions of Triethyl Phosphite.

ACS omega·2026

Same author

Genome-wide association study and transcriptome analysis identify candidate genes associated with low nitrogen-induced root plasticity in Zea mays L.

Annals of botany·2026

Same author

Genetic Basis of Nitrogen-Deficiency-Induced Root Cortical Aerenchyma in Maize Revealed by GWAS and Transcriptome Analysis.

Plants (Basel, Switzerland)·2026

Same author

Effects of computerized cognitive training on brain function in children with ADHD: A longitudinal neuroimaging study based on fALFF.

Behavioural brain research·2025

Same journal

Tissue MicroRNAs in Arrhythmogenic Cardiomyopathy: A Systematic Review of Studies in Human Myocardium and Animal Models with Implications for Post-Mortem Molecular Diagnostics.

Genes·2026

Same journal

Genetic Variants and Dental Caries Susceptibility: An Umbrella Review and Multilevel Meta-Analysis.

Genes·2026

Same journal

Generative AI and Language Models in Human Genetics and Health: From Variant Interpretation to Clinical Decision Support.

Genes·2026

Same journal

Familial White-Sutton Syndrome Caused by a Pathogenic POGZ p.Arg508* Variant: Intrafamilial Variability from Childhood to Adulthood.

Genes·2026

Same journal

Genetic Influence on LDL-Cholesterol Levels: Role of Polygenic Risk Scores and Lp(a) Beyond Monogenic Hypercholesterolemia.

Genes·2026

Same journal

THBS1 as a Key Regulator of Myoblasts: Validation of Its Inhibitory Roles in Skeletal Muscle Development.

Genes·2026

See all related articles

CMSV is a new deep learning method for detecting structural variations in long-read sequencing data. It accurately identifies various structural variants, improving genomic diversity and disease research.

Area of Science:

Genomics
Bioinformatics
Computational Biology

Background:

Structural variations (SVs) are key genomic variations linked to human diseases.
Long-read sequencing aids SV detection, but current methods struggle with complex regions and joint modeling.
Existing heuristic rules limit the accuracy of structural variation detection.

Purpose of the Study:

Introduce CMSV, a novel method for structural variation detection and genotyping using long-read sequencing data.
Address limitations in modeling local SV signatures and cross-subsegment context.
Provide a robust framework for diverse structural variation types.

Main Methods:

Utilize multi-channel, position-level features from alignment data.
Employ a multi-scale convolutional encoder with stacked Mamba modules for candidate region detection.

Keywords:

Mamba SV detection deep learning long-read sequencing structural variation

Related Experiment Videos

Integrate DBSCAN and length-based clustering for variant optimization and genotype inference.

Main Results:

CMSV demonstrates competitive performance across PacBio and ONT platforms for DEL/INS detection and genotyping.
Validated ability to detect DUP, INV, and TRA/BND using simulated multi-type datasets.
Exhibits stable performance and good family-level consistency in trio-based evaluations.

Conclusions:

CMSV offers an effective deep learning framework for long-read SV detection and genotyping.
The method is versatile across different sequencing platforms and coverage levels.
Enhances the study of genomic diversity and disease associations through improved SV analysis.