For each sequence S i, we calculate the relative frequencies of all possible spaced words with respect to our pattern P (relative to the sequence length) and, similar as in other alignment-free approaches, each sequence S i is represented by the -dimensional vector of these relative frequencies.
The validated targets used to create the alignment showed a greater degree of heterogeneity that those in other alignments.
However, in one alignment, codons are conserved (same codons occur in the same column), while in the other alignment codons appear only once in a column.
Alignments with low scores are typically pairs with low frequency where the source and target words occur in many other alignment pairs.
The logos produced by Spial appear similar to those produced by the program Two Sample Logo (Vacic et al., 2006), which treats one alignment as the background and then computes whether there are residues that are enriched or depleted in the other alignment.
The benchmarking of PINALOG in comparison with other alignment methods shows that PINALOG obtains a good balance in including these features as it generates an alignment with consistently high number of conserved interactions, homologous aligned pairs and functionally similar protein pairs.
DIALIGN has been shown to perform well relative to other alignment programs in aligning conserved domains within rapidly evolving (with respect to both indels and substitutions) regions [ 53, 58, 59].
In contrast to other alignment methods proposed for PET, which suffer from potentially worse performance and efficiency, track-based alignment in general is based on the minimization of residuals by exploiting the knowledge of source positions (known with high accuracy).
The mRNA-to-genome and trimmed probe set consensus sequence alignments were then sorted into groups such that at least one alignment block from each group member overlapped with one or more blocks from at least one other alignment in the same group.
