The genome sequences of new viruses often contain many “orphan” or “taxon-specific” proteins apparently lacking homologs. However, because viral proteins evolve very fast, commonly used sequence similarity detection methods such as BLAST may overlook homologs. We analyzed a data set of proteins from RNA viruses characterized as “genus specific” by BLAST. More powerful methods developed recently, such as HHblits or HHpred (available through web-based, user-friendly interfaces), could detect distant homologs of a quarter of these proteins, suggesting that these methods should be used to annotate viral genomes. In-depth manual analyses of a subset of the remaining sequences, guided by contextual information such as taxonomy, gene order, or domain cooccurrence, identified distant homologs of another third. Thus, a combination of powerful automated methods and manual analyses can uncover distant homologs of many proteins thought to be orphans. We expect these methodological results to be also applicable to cellular organisms, since they generally evolve much more slowly than RNA viruses. As an application, we reanalyzed the genome of a bee pathogen, Chronic bee paralysis virus (CBPV). We could identify homologs of most of its proteins thought to be orphans; in each case, identifying homologs provided functional clues. We discovered that CBPV encodes a domain homologous to the Alphavirus methyltransferase-guanylyltransferase; a putative membrane protein, SP24, with homologs in unrelated insect viruses and insect-transmitted plant viruses having different morphologies (cileviruses, higreviruses, blunerviruses, negeviruses); and a putative virion glycoprotein, ORF2, also found in negeviruses. SP24 and ORF2 are probably major structural components of the virions.

As biomolecular sequencing is becoming the main technique in life sciences, functional interpretation of sequences in terms of biomolecular mechanisms with in silico approaches is getting increasingly significant. Function prediction tools are most powerful for protein-coding sequences; yet, the concepts and technologies used for this purpose are not well reflected in bioinformatics textbooks. Notably, protein sequences typically consist of globular domains and non-globular segments. The two types of regions require cardinally different approaches for function prediction. Whereas the former are classic targets for homology-inspired function transfer based on remnant, yet statistically significant sequence similarity to other, characterized sequences, the latter type of regions are characterized by compositional bias or simple, repetitive patterns and require lexical analysis and/or empirical sequence pattern-function correlations. The recipe for function prediction recommends first to find all types of non-globular segments and, then, to subject the remaining query sequence to sequence similarity searches. We provide an updated description of the ANNOTATOR software environment as an advanced example of a software platform that facilitates protein sequence-based function prediction.

Summary: The usage of current sequence search tools becomes increasingly slower as databases of protein sequences continue to grow exponentially. Tachyon, a new algorithm that identifies closely related protein sequences ~200 times faster than standard BLAST, circumvents this limitation with a reduced database and oligopeptide matching heuristic.Availability and implementation: The tool is publicly accessible as a webserver at http://tachyon.bii.a-star.edu.sg and can also be accessed programmatically through SOAP.Contact: sebastianms@bii.a-star.edu.sgSupplementary information: Supplementary data are available at the Bioinformatics online.