To get more accurate alignments, we ran MAFFT with L-INS-i iterative refinement method, which is best suited for aligning sequences with one conserved domain and long gaps.
Moreover, McGowen et al. used the program Muscle to align the genes, whereas we used Prank, a program that is reported to be a much more accurate aligner (Fletcher and Yang 2010).
For instance, the combination of RNAplfold (Bernhart et al., 2006) predicted secondary structures made R-Coffee more accurate at aligning RNA sequences than any of the alternative regular aligners, with a 4-point net improvement as estimated on BraliBase (Wilm et al., 2008).
First, because DACTAL's decomposition strategy produces subsets with at most 250 sequences, it is possible to run MAFFT in more accurate ways to align each subset than may be possible on the full taxon set.
Composition-based statistics use the composition of the sequences being aligned to generate more accurate E-values, which allows for a more accurate distinction between true and false matches.
To obtain a more accurate phylogenetic assignment of the OTUs, the aligned sequences were divided into phyla and, in the case of Proteobacteria, into classes.
The top 40 hits for each consensus were extracted, this time with 1,000-bp 1,000-bp sequence, and aligned to produce a more accurate consensus.
Identification of peaks from aligned DASH traces is more accurate, which is essential for identification of peaks which migrate closely together.
DNA size selection allowed us to choose an appropriate library insert length with precision, such that mate-pair sequences of each insert overlap in the middle, and can be aligned to produce a longer, more accurate composite read.
Given short and error-prone reads, SSAHA2 is more accurate, although it has to spend more time on aligning such reads.
