Customized de novo mutation detection for any variant calling pipeline: SynthDNM

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Advancements in sequencing technologies necessitate adaptable de novo mutation (DNM) calling tools.
• Accurate identification of DNMs from diverse genome and exome datasets is crucial.

Purpose of the Study:

• To develop a flexible and accurate de novo mutation calling classifier.
• To create a tool adaptable to evolving sequencing and variant calling pipelines.

Main Methods:

• Developed SynthDNM, a random-forest based classifier.
• Employed a flexible approach using simulated training examples derived from real data.
• Optimized classifiers for predicting de novo single nucleotide polymorphisms (SNPs) and indels.

Main Results:

• SynthDNM demonstrates robust accuracy in predicting de novo SNPs and indels.
• The classifier performs well across multiple variant calling methods.
• SynthDNM is readily adaptable to new sequencing and variant-calling pipelines.

Conclusions:

• SynthDNM offers a flexible and accurate solution for de novo mutation calling.
• The tool supports the analysis of diverse genomic datasets and pipelines.
• SynthDNM is freely available for use and adaptation.