source: trunk/HELP_SOURCE/source/awt_csp.hlp

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

#       sub topics:
SUB     pos_var_pars.hlp

# format described in ../help.readme


TITLE           Estimate Parameters from Column Statistics

OCCURRENCE      ARB_DIST

DESCRIPTION     In a standard RNA, base frequencies are not equally
                distributed. Especially in the archea subclass we find
                extremely G+C rich sequences.
                This yielded in a couple of new rate corrections, algorithms
                and programs which:

                        - calculate the average G+C content of all/two sequences
                        - correct the distance.

                But further research showed us that the G+C frequencies are
                not equally distributed within a sequence. Especially helical
                parts have a significant higher G+C content than non
                helical parts.
                One strait forward algorithm would calculate each frequency
                independently for each column.
                Especially for small datasets the resulting frequencies would
                look like random data, as too few examples are analyzed.

                In ARB we implemented a combination of the 2 approaches.
                        Lets say we want to estimate a Parameter 'P' with
                        a maximum variance 'maxvar', so we need a minimum
                        samples 'minsap'.

                        - All sequence positions are clustered according to

                                - helical/non helical region
                                - variability

                          The size of the cluster is choosen with respect
                          to the variability of the sequences to get a
                          minimum of independent events.

                        - The final parameter estimate for a column is a
                          weighted sum between the estimate for the
                          cluster and the estimate for the single position.

                You can give your favorite method a higher weight by
                controlling the smoothing parameter:

                        Less smoothing -> independent parameter estimates

                        Much smoothing -> clustered parameter estimates

                To get a good tree we recommend you to try all selections.

NOTES           To get parameters from a column statistic you first have
                to create one.
                Do this with <ARB/SAI/Create SAI using../Positional variability.. (parsimony method)>

WARNINGS        Problems may occur when

                         1. 'independent parameter estimates' is selected and
                         2. your dataset is quite small (<100 Sequences) and
                         3. one sequence is bad or badly aligned

                or

                         1. Much smoothing of parameters is selected and
                         2. you are analyzing ribosomal RNA and
                         3. 'Use Helix Information' is turned off


BUGS            No bugs known