ADTEx

ADTEx detects copy number variations (CNVs) and genotypes from paired tumor and matched normal whole-exome sequencing data to characterize somatic genomic alterations in cancer.

Key Features:

• Dual Hidden Markov Models: Employs two Hidden Markov Models to infer copy number and genotypes from exome data by analyzing depth of coverage ratios and B allele frequencies.
• Resolution of complex genomic events: Accounts for polyploidy, aneuploidy, normal cell contamination, and signal baseline shifts to improve CNV detection accuracy.
• Chromosome arm-level detection: Explicitly detects CNVs at the chromosome arm level commonly observed in tumor genomes.
• Performance and validation: Evaluated on 17 in-house and 18 TCGA paired ovarian cancer/normal exomes and benchmarked against CNVs and genotypes predicted using Affymetrix SNP 6.0, reporting superior precision and F-measure compared to competitors.
• Implementation: Implemented using Python and R.

Scientific Applications:

• Cancer genomics: Detection and characterization of somatic CNVs in tumor samples using whole-exome sequencing.
• Study of tumorigenesis and progression: Identification of arm-level and focal genomic alterations associated with cancer progression.
• Translational research: Supporting investigations that relate somatic CNVs to potential personalized treatment strategies and cancer genetics.

Methodology:

ADTEx uses two Hidden Markov Models to infer copy number states and genotypes from depth of coverage ratios and B allele frequencies in paired tumor-normal whole-exome sequencing data, detecting chromosome-arm level events and accounting for polyploidy, aneuploidy, normal cell contamination, and signal baseline shifts; it is implemented in Python and R.