appreci8: a pipeline for precise variant calling integrating 8 tools.

Area of Science:

• Genomics
• Bioinformatics
• Clinical Diagnostics

Background:

• Next-generation sequencing (NGS) requires accurate variant calling for clinical applications.
• Current variant calling tools exhibit limitations, producing false positives and negatives.
• This necessitates manual validation and additional experiments, increasing costs and time.

Purpose of the Study:

• To develop an automated variant calling pipeline to improve accuracy and efficiency.
• To combine and filter outputs from multiple open-source variant callers.
• To establish a robust method for single nucleotide variant and short indel detection.

Main Methods:

• Developed appreci8, an automated variant calling pipeline.
• Integrated and filtered outputs from eight open-source variant calling tools.
• Utilized a novel artifact- and polymorphism score for training and validation.
• Tested on diverse datasets covering various sequencing platforms, target regions, and disease entities.

Main Results:

• Appreci8 demonstrated high sensitivity (0.93-1.00) and positive predictive value (0.65-1.00).
• The pipeline showed superior performance compared to individual tools and alternative approaches.
• Validated calls through re-sequencing and expert review across 678 patient samples.
• Achieved accurate variant calling for single nucleotide variants and short indels.

Conclusions:

• Appreci8 offers a significant improvement in variant calling accuracy for clinical NGS.
• The automated pipeline reduces the need for extensive manual work and additional experiments.
• Appreci8 provides a reliable and efficient solution for variant detection in research and clinical settings.