transalign

transalign constructs transitive pairwise sequence alignments by combining alignments against large uncurated databases and curated target subsets to improve detection of homologous sequences.

Key Features:

• Indirect alignment methodology: Aligns query sequences first to a comprehensive, large uncurated database to capture potential homologs that may be absent or weak in smaller curated subsets.
• Transitive alignments: Constructs indirect alignments by integrating results from the large-database alignments with alignments between the large database and the target curated subset.
• Increased sensitivity: Enhances sensitivity of pairwise sequence alignments to detect homologous relationships when direct hits in the target database are sparse or weak.
• Mitigation of random-hit ambiguity: Improves discrimination between true homologous sequences and random hits arising from limited representation in curated databases.

Scientific Applications:

• Curated-subset sequence retrieval: Identifying sequences with specific characteristics from a curated subset, including sequences from a particular species, with known 3D structures, or with reliable functional annotations.
• Comparative genomics: Enhancing detection of homologs across genomes by leveraging broader sequence space to inform alignments within curated sets.
• Evolutionary biology: Improving inference of evolutionary relationships by increasing sensitivity to distant homologs through transitive alignment information.
• Functional annotation of novel sequences: Aiding annotation by connecting novel queries to annotated sequences in curated databases via the larger sequence corpus.

Methodology:

Query sequences are first aligned against a large, uncurated database; alignments from that phase are then combined with alignments between the large database and the target curated subset to construct indirect (transitive) pairwise alignments.