maxAlike: Sequence Reconstruction by Maximum Likelihood Estimation

The task of reconstructing a piece of sequence in a particular species is gaining more and more importance in the light of the rapid development of high-throughput sequencing technologies and their limitations. Applications include not only compensation for missing data in unsequenced genomic regions but also the preparation of customized queries for homology-based searches. Here, we introduce the maxAlike web server. It takes a multiple sequence alignment and a phylogenetic tree that also contains a target species as input. For the target species, it computes nucleotide probabilities as well as the most likely sequence, which can be used for primer design or homology search. Furthermore, position specific scoring matrices (PSSMs) of regions of high confidence are available for download. We show that as much as 99% of a sequence can be reconstructed correctly using the maxAlike algorithm, when the sequence of a closely related species is available, compared to only 89% reconstructed positions using only the consensus sequence from the input alignment. For more distant species, the reconstruction rate of maxAlike drops to a plateau value of about 60–70% for the maxAlike approach, compared to 50–60% for the consensus sequence. The web server is freely accessible at: http://rth.ku.dk/resources/maxAlike.