We collected all motifs of 6 to 12 base pair length from ungapped (upper case) alignment columns.
In the latter case alignment information (position and length of gaps) can be imported back into the annotated dataset in ProSeq3.
In these cases, alignments will be poor and phylogenetic analyses may be prone to long-branch attraction.
For uncertain cases, alignments with T-Coffee [39] and different ClustalW parameters were compared to combine them into a final alignment.
Nevertheless, it seems that in most cases, alignments with higher Alignment Scores have better BALiBASE score as well.
In other cases, alignments revealed the presence of large inserts within nucleotide sequences, likely corresponding to unspliced introns.
Apart from these cases, alignments with low π values were those of ribosomal proteins, histones and cytochromes among others.
The "quality" of the alignment algorithms, i.e. mutual concordance of algorithmic and GS alignments, was analyzed from different points of view; in most cases, alignments based on intercomparison of three-dimensional structures were used as the GS alignments.
Because in some cases alignments might contain sequences that are shorter at one or both ends, the effective length of the alignment (ignoring regions covered by a single sequence) was used to compute alignment attributes, such as the SNP density.
