There are also several examples of low complexity words (Supplementary Material Table 7), similar to Y. lipoloytica.
The fungus Y. lipolytica has only low complexity words within its major vocabulary (Supplementary Material Table 6).
Low complexity words, such as 'ATGC'MEPQSDVLDVL', or '0123456789', are considered as unique sets of symbols and are not considered in the vectors.
The set SIG1 is computed as described in [ 42, 43], which is summarized as follows: 1. Identify maximally repeated words of length [m, n]. 2. Remove low complexity words, redundant words, and words that are contained in repeat elements.
Rare sequencing differences were identified via BlastN (low complexity filter off, word size 7) [ 49] by applying the SMRT-derived genome sequence as a reference.
In order to make as long alignments as possible, especially for those proteins with very low sequence identities, we disabled the filter for low complexity regions, set the word size to be 2 and chose the highest E-value cutoff available for both tools.
GS-FLX reads were mapped to reference sequences using Blast v. 2.2.15 [ 23], with e value set to 1e-7; low complexity filter (DUST) disabled; word size, 4; and -v and -b set to the highest possible number of matches (in a dataset containing 100 000 reads, -v and -b are set to 100 000 or higher).
To increase the sensitivity of the blastn search, the option of filtering low complexity regions was disabled, the word size was set to two nucleotides, and the expected e-value was relaxed to 1E+3.
Each of the five β-amylase elements was used for a separate nucleotide-nucleotide BLAST search of all organisms in Entrez, with low complexity filter on, and a word size of 11.
We conducted some further tests to boost confidence in these single peptide identifications, searching them back against the genome sequence using tblastn (with no low complexity filtering, and expect value, word size, and database size parameters optimised for short matches).
funestus conceptual proteins were gathered into a database that was searched by BLASTP with each component protein, with the filter for low complexity set to False (-F F), word size = 2 (-W 2) and using the default compositional matrix adjustment ON.
