JESAM

JESAM performs clustering and alignment of Expressed Sequence Tags (ESTs) to identify overlaps, reduce redundancy, and support analysis of eukaryotic gene expression from cDNA-derived sequences.

Key Features:

• Error Reduction and Redundancy Management: Generates derived database views of public EST submissions to mitigate sequencing errors and high redundancy in EST databases.
• Supporting Evidence Publication: Publishes supporting evidence and intermediary results for EST clusters and consensus to enable verification and customization of outputs.
• Efficient Intermediary Calculations: Identifies and publishes resource-intensive common calculations used across EST analysis methods to minimize redundant computation.
• Parallel Processing with PVM: Performs sequence comparisons in parallel using a Parallel Virtual Machine (PVM) process farm for scalable processing of large datasets.
• Incremental Updates: Stores previous results to enable incremental updates without reprocessing entire datasets.
• Global Access via CORBA Interfaces: Exports overlap databases through Common Object Request Broker Architecture (CORBA) interfaces for programmatic global access.
• Comparison and Contrast Capabilities: Compares outputs with established databases such as UniGene Mouse and Rat to highlight differences and undesirable features.
• Software Packaging: Distributes code in two Java Archive (JAR) files: one containing Java source and another containing C, C++, and Interface Definition Language (IDL) code.

Scientific Applications:

• Mapping Sequence Tag Sites (STSs): Facilitates creation of genetic maps by identifying sequence tag sites (STSs).
• Polymorphism Discovery: Assists identification of genetic polymorphisms from EST overlaps and clusters.
• Disease Gene Hunting: Supports discovery of disease-associated genes by aggregating EST evidence.
• Proteomics Integration: Provides EST-derived sequence data to support mass spectrometer proteomics for protein identification.

Methodology:

Self-compares databases of EST and full-length mRNA sequences to identify overlaps consistent with contiguity, executes comparisons in parallel using a PVM process farm, and stores results incrementally to allow updates.