VarScan is an open-source tool designed for variant detection in massively parallel sequencing data, compatible with multiple short-read aligners such as BLAT, Newbler, cross_match, Bowtie, and Novoalign. It can detect SNPs and indels with high sensitivity and specificity in both individual and pooled samples. The tool processes alignments by scoring and sorting reads, discarding those with low identity or multiple alignments, and then identifying sequence changes in the best alignment for each read. VarScan reports variants with their chromosomal coordinates, alleles, flanking sequences, and read counts. Performance evaluations using real data from Roche/454 and Illumina/Solexa platforms showed high sensitivity and specificity, with 97% specificity in individual 454 data and 93% sensitivity in pooled Illumina data. VarScan's ability to detect low-frequency variants and its platform-independence make it a powerful tool for large-scale targeted studies of genetic variation.VarScan is an open-source tool designed for variant detection in massively parallel sequencing data, compatible with multiple short-read aligners such as BLAT, Newbler, cross_match, Bowtie, and Novoalign. It can detect SNPs and indels with high sensitivity and specificity in both individual and pooled samples. The tool processes alignments by scoring and sorting reads, discarding those with low identity or multiple alignments, and then identifying sequence changes in the best alignment for each read. VarScan reports variants with their chromosomal coordinates, alleles, flanking sequences, and read counts. Performance evaluations using real data from Roche/454 and Illumina/Solexa platforms showed high sensitivity and specificity, with 97% specificity in individual 454 data and 93% sensitivity in pooled Illumina data. VarScan's ability to detect low-frequency variants and its platform-independence make it a powerful tool for large-scale targeted studies of genetic variation.