source: branches/lib/HELP_SOURCE/source/phylo.hlp

#       main topics:
UP      arb.hlp
UP      glossary.hlp

#       sub topics:
#SUB     subtopic.hlp

# format described in ../help.readme


TITLE           ARB_PHYLO: Create filter by base frequency

OCCURRENCE      ARB_NT/SAI/Create Filter/by Base Frequency

DESCRIPTION     Calculate base frequencies and/or a base frequency filter.

                Species scope:

                    The base frequencies will be calculated upon those species
                    that are marked when ARB_PHYLO is started up (changing marks
                    later has no effect).

                Open Config/Filter to define how the filter will be created:

                     Define the alignment range, for which the filter shall
                     be generated using the input fields 'Start at column'
                     and 'Stop at column'. Columns outside that range will get
                     filtered, i.e. the generated filter will contain '0'.

                     Define the similarity range, which will lead to a '1' in the
                     resulting base frequency filter, by specifying the lower and
                     upper percentage in the input field 'Minimal similarity'
                     and 'Maximal similarity'. Similarity values outside that range
                     will lead to '0' in the generated base frequency filter and will
                     as well not be exported as base frequencies. (Note: It's not necessary to
                     re-calculate, when you change this value)

                     For non-regular sequence characters (gaps, ambiguity codes and lowercase characters)
                     you may define how the frequency calculation acts, by setting their toggles to one of
                     the following values:

                        don't count (ignore)

                              The corresponding characters are ignored (as if they were not present).

                        if occurs most often => forget whole column

                              If the corresponding character(s) occur(s) more often than any other
                              regular sequence character, then skip column, i.e. write '0' to filter.

                        if occurs => forget whole column

                              If (one of) the corresponding character(s) occurs, then skip the column,
                              i.e. write '0' to filter.

                        count, but do NOT use as maximum

                              The corresponding characters are counted like regular sequence characters,
                              but never interpreted as "being the most homologous character".
                              Compared with 'don't count' it will result in lower base frequencies, cause
                              the overall number of characters occurring per column will be higher.

                        treat as regular character

                              This is only applicable to 'ambiguity codes'. Treats ambiguity codes like
                              regular sequence characters. (Note: 'treat as uppercase char' setting for
                              lowercase characters affects ambiguity codes as well, if treated as regular
                              characters)

                        treat as uppercase char

                              This is only applicable to 'lowercase chars'. Simply treat regular lowercase
                              sequence characters like uppercase characters.

                Press 'Calculate/Column Filter' to calculate the base frequencies.

                When calculated, three lines will appear below your sequences in the main window.
                Reading the columns of these 3 lines from top to bottom, they mark completely
                filtered columns by 'XXX' and else show the frequency of the most frequent base in percent.

                The color of each column indicates whether the frequency will be exported and respectively
                whether the column will be '0' or '1' in the base frequency filter.

                If pleased with the results, either export detailed base frequencies
                via 'File/Export frequencies' or export a base frequency filter via 'File/Export filter'.

NOTES           None

EXAMPLES        To create a filter which hits all columns containing sequence data, use the following settings:

                   - all alignment columns
                   - similarity 0 to 100
                   - both gaps: 'don't count (ignore)'
                   - ambiguity codes: 'treat as regular character'
                   - lowercase chars: 'treat as uppercase char'

WARNINGS        None

BUGS            No bugs known