GMAP

GMAP maps and aligns cDNA, mRNA, and expressed sequence tag (EST) sequences to genomic DNA to identify exon–intron structures and splice sites for gene-structure inference and annotation.

Key Features:

• Efficient mapping and alignment: Maps and aligns mRNA and expressed sequence tag (EST) sequences to a genome without relying on probabilistic splice-site models.
• Accurate gene-structure inference: Generates precise gene structures despite substantial polymorphisms and sequence errors.
• Robust alignment strategies: Employs a minimal sampling strategy for genomic mapping, oligomer chaining for approximate alignment, and sandwich dynamic programming (DP) for splice-site detection.
• Microexon identification: Detects microexons and evaluates them using statistical significance testing.
• Empirical accuracy on mutated sequences: In tests on human mRNAs with random 1% and 3% mutations, it identified all splice sites in over 99.3% of sequences.
• Comparative performance: Provided higher-quality alignments more frequently than BLAT on large human EST sets and performed comparably to GeneSeqer on Arabidopsis cDNAs.
• Speed and resource efficiency: Achieves several-fold faster processing with low startup time and memory requirements suitable for high-throughput applications.

Scientific Applications:

• Gene structure prediction and annotation: Infers exon–intron boundaries and gene models from cDNA, mRNA, and EST alignments.
• Comparative genomics: Aligns mRNA and EST sequences for comparative analyses across genomes.
• Polymorphism and error analysis: Analyzes effects of polymorphisms and sequence errors on inferred gene structures.
• High-throughput sequencing projects: Supports alignment of large batches of cDNA/mRNA/EST sequences in high-throughput workflows.

Methodology:

Maps and aligns cDNA (including mRNA and EST) sequences to genomic DNA using a minimal sampling strategy for genomic mapping, oligomer chaining for approximate alignment, sandwich dynamic programming (DP) for splice-site detection, and microexon identification with statistical significance testing, without relying on probabilistic splice-site models.