Alternative splicing prediction database (ASPICdb)

Alternative splicing prediction database (ASPICdb) provides annotated alternative splicing patterns in human genes by applying the ASPic and Pintron algorithms to identify splice sites, predict full-length mRNA isoforms, and infer gene structures.

Key Features:

• Alternative splicing annotation: Uses the ASPic algorithm to identify splice sites and annotate alternative splicing patterns from transcript and genome data.
• Gene structure prediction (Pintron): Uses the Pintron algorithm to compute spliced alignments, exploit redundancy in transcript clusters, and predict exon-intron gene structures.
• Integration with Ensembl and Unigene: Cross-references predicted isoforms and gene annotations with Ensembl and Unigene.
• Functional annotation of isoforms: Provides functional annotations for predicted mRNA and protein isoforms.
• Comparative analysis across species: Performs comparative analysis of transcript and genome data across multiple species to support evolutionary and functional inference.
• Scalability for large transcript datasets: Processes large clusters of transcript sequences using Pintron algorithms and spliced-alignment methods.

Scientific Applications:

• Transcriptome and proteome complexity analysis: Elucidates how single gene loci produce multiple mRNA and protein isoforms to analyze transcriptomic and proteomic diversity.
• Gene structure prediction: Predicts exon and intron organization from transcript clusters using Pintron-derived spliced alignments.
• Comparative genomics studies: Compares splicing patterns across species via ASPic comparative analysis to inform evolutionary and functional genomics.
• Study of gene expression regulation: Supports investigation of regulation mediated by alternative splicing through annotated splice sites and isoforms.

Methodology:

ASPICdb applies the ASPic algorithm for comparative analysis of transcript and genome data across multiple species to identify splice sites and predict compatible mRNA isoforms, and applies the Pintron algorithm/package to compute spliced alignments, exploit redundancy in transcript clusters, and predict exon-intron gene structures from large transcript datasets.