Maximum base frequency

ARB_NT/SAI/create SAI/Max Frequency

Finds the most frequent base (or gap) in each column for all marked species. Then the number of all sequences with this base are divided by:

• the number of all marked sequences (if not ignoring gaps)
• the number of bases in this column (if ignoring gaps)

The resulting percentage is divided by ten and then the second last digit is taken:

0%  - 19%  ->  '1' (does not occur for nucleotides)
20% - 29%  ->  '2'
30% - 39%  ->  '3'
...
90% - 99%  ->  '9'
100%       ->  '0'

The result can be used as a conservation profile and filter. Rule of thumb: the higher the number, the more conserved the position (but mind the '0' which means 100%!).

Internally the SAI consists of two lines: the main line called 'data' and a second line called 'dat2'.

Say one column contains 7 A's 4 G's and 5 Gaps.

• ignoring gaps will result in 7/11 == 64 % which is converted to '6'.
• otherwise we get 7/16 == 44% which will be indicated by a '4' in the target sequence.

If gaps are ignored '-' are treated like '.': both get removed and frequency is calculated on non-gaps only.

If gaps are NOT ignored, '-' are treated like non-gaps, i.e. a column containing only '-' will be assigned a max. frequency of 100%. '.' are treated as gaps.

Ambiguities are counted proportionally, i.e.

• a 'N' counts as 1/4 'A', 1/4 'C', 1/4 'G' and 1/4 'T'
• a 'D' counts as 1/3 'A', 1/3 'G' and 1/3 'T'
• a 'Y' counts as 1/2 'C' and 1/2 'T'

Example:

No bugs known yet