GSNAP is a fast and SNP-tolerant program for detecting complex variants and splicing in short reads. It aligns single- and paired-end reads as short as 14 nt and can detect short- and long-distance splicing, including interchromosomal splicing, using probabilistic models or a database of known splice sites. It also allows SNP-tolerant alignment to a reference space of all possible combinations of major and minor alleles, and can align reads from bisulfite-treated DNA for methylation studies. GSNAP uses a successively constrained search process of merging and filtering position lists from a genomic index to detect complex variants and splicing. It can detect splicing using probabilistic models or known splice sites, and can identify splicing events with mismatches, partial splicing, or 'half intron' events. It can also detect long-distance intrachromosomal deletions, inversions, and interchromosomal translocations. GSNAP can align reads to a reference space of all possible combinations of major and minor alleles, allowing it to avoid treating minor alleles as mismatches. This SNP-tolerant alignment can reveal alternate genomic mappings for a read, which is particularly useful when a read fails to align due to two minor alleles nearby. GSNAP is efficient in detecting complex variants with four or more mismatches or insertions of 1–9 nt and deletions of 1–30 nt, especially for reads of 70 nt or longer. It is also efficient in detecting splicing events, including interchromosomal splicing. It can align reads from bisulfite-treated DNA for methylation studies, and can detect mismatches between genomic-T and read-C, which are obscured by C–T substitutions. GSNAP uses a multiway merging process to generate and filter candidate genomic regions, and can detect complex variants and splicing with high sensitivity and speed. It can detect insertions and deletions, and can detect splice junctions using known or novel splice sites. It can also align paired-end reads, and can detect long-distance splicing by identifying known or novel splice sites. GSNAP has been tested against other alignment programs and has shown comparable or faster performance in detecting complex variants and splicing. It has also shown high sensitivity in detecting intragenic and intergenic exon–exon junctions, and can detect splicing events with high accuracy. It has also shown high alignment yields for transcriptional reads, with SNP tolerance increasing alignment yields by up to 8–9%.GSNAP is a fast and SNP-tolerant program for detecting complex variants and splicing in short reads. It aligns single- and paired-end reads as short as 14 nt and can detect short- and long-distance splicing, including interchromosomal splicing, using probabilistic models or a database of known splice sites. It also allows SNP-tolerant alignment to a reference space of all possible combinations of major and minor alleles, and can align reads from bisulfite-treated DNA for methylation studies. GSNAP uses a successively constrained search process of merging and filtering position lists from a genomic index to detect complex variants and splicing. It can detect splicing using probabilistic models or known splice sites, and can identify splicing events with mismatches, partial splicing, or 'half intron' events. It can also detect long-distance intrachromosomal deletions, inversions, and interchromosomal translocations. GSNAP can align reads to a reference space of all possible combinations of major and minor alleles, allowing it to avoid treating minor alleles as mismatches. This SNP-tolerant alignment can reveal alternate genomic mappings for a read, which is particularly useful when a read fails to align due to two minor alleles nearby. GSNAP is efficient in detecting complex variants with four or more mismatches or insertions of 1–9 nt and deletions of 1–30 nt, especially for reads of 70 nt or longer. It is also efficient in detecting splicing events, including interchromosomal splicing. It can align reads from bisulfite-treated DNA for methylation studies, and can detect mismatches between genomic-T and read-C, which are obscured by C–T substitutions. GSNAP uses a multiway merging process to generate and filter candidate genomic regions, and can detect complex variants and splicing with high sensitivity and speed. It can detect insertions and deletions, and can detect splice junctions using known or novel splice sites. It can also align paired-end reads, and can detect long-distance splicing by identifying known or novel splice sites. GSNAP has been tested against other alignment programs and has shown comparable or faster performance in detecting complex variants and splicing. It has also shown high sensitivity in detecting intragenic and intergenic exon–exon junctions, and can detect splicing events with high accuracy. It has also shown high alignment yields for transcriptional reads, with SNP tolerance increasing alignment yields by up to 8–9%.