DNMFilter_Indel

DNMFilter_Indel filters false positive de novo insertions and deletions (indels) from whole-genome and exome sequencing data of parent-offspring trios to improve accuracy of de novo indel discovery for genetic research and clinical genetics.

Key Features:

• Machine Learning Approach: Employs a gradient boosting algorithm to distinguish true de novo indels from false positives.
• Compatibility and Integration: Integrates with other de novo indel detection tools to increase specificity of candidate indel calls.
• Sequencing Data Support: Operates on whole-genome and whole-exome sequencing data from parent-offspring trios.
• Performance Metrics: Demonstrated on real genome sequencing data to substantially reduce false positive rates while maintaining high sensitivity.
• Implementation: Implemented using Java and R.

Scientific Applications:

• Genetic Research: Enables more accurate investigation of the roles of de novo indels in genetic disorders by reducing false positives in trio sequencing analyses.
• Clinical Genetics: Improves precision of de novo indel detection for diagnosis and personalized treatment planning.

Methodology:

Applies a gradient boosting classification algorithm for candidate de novo indel filtering; software implemented in Java and R.