CVS log for ccp4/src/peakmax.f
	Help

Request diff between arbitrary revisions / Display revisions graphically

XML: write messages as well

update licence banners

Info on possible translational NCS was not being output

Added EXCLUDE EDGE|NONE option (turns suppression of peaks at the edge of
the map on or off) (see bug #1090).

add new licence banners

changes from Clemens

set explicit widths for internal io

search top 6 peaks for NCS

Apparently the use of internal list-directed writes with no external
list-directed writes causes problems for gcc3.1 on the mac,
therefore, list-directed writes switched to formatted

PRE: fix for OUTPUT PDB option, which was misaligning the atomname in the
output file.

PRE: fix for OUTPUT PDB option, which was misaligning the atomname in the
output file.

CV- declorations

CV - declorations

formatting change from EJD

output formatting change from EJD

trailing spaces (CV)

 I have fixed peakmax to flag NCS translation - needed for MR, but only
if the PATT keyword is given - ejd

optionally produce XML file with top peaks

Expanded field width when printing out peak heights.

Defaults changed: ATNAME=O, RESIDUE=HOH.

add pdb for tact - not just bookhaven! ejd

 Clemens Vonrhein fix to bfactor card

Minor correction from EJD for OUTPUT FRAC.

Last minute changes from Liz.

g77 complaint about non exsistant s/r in EXTERNAL statement

Removed s/r MATVEC2 and replaced with call to library s/r MATMULGEN.

1. New output option FRAC added.
2. Output PEAKS option now writes to logical name PEAKS.
3. OUTPUT option fixed (was partially broken).
4. Program now outputs the height/rms density instead of height in output
   BROOK and in logfile. Some formats changed in logfile.
5. Renamed MATVEC as MATVEC2 to avoid name clash with routine in modlib.

peakmax doesnt write out the matrix at the head of the BROOKHAVEN output..

Yet more changes to RWBROOK.

adde new rwbrook stuff

Added extra check on dimension limits.

files CLOSED.

Kim Henrick: add keyword PATTERSON which then gives vector lengths on
sorted list.

192 symops

Rejected symmetry related peaks at map edge are not displayed in sorted peaks.

Removed output overflow.

IJT: Previous fix rejected second redundant peak; however this is not
necessarily the higher one.  Fix below rejects lower (absolute) of pair.
This only matters if the output file is further processed and a threshold
is applied (which is what I'm doing).

tidying

IJT: stop symmetry related peaks being output

replace rwatom calls with rwbrook

merge back undef var change

The incorrect peak height was being written to the
output Brookhaven file in the "occupancy" column. The height
of the NEXT peak in the sorted list rather than the current
peak was output. Thus the output Brookhaven file could have a
negative occupancy but a positive BTEMP ! (Andrew)

undefined vars

version pulled straight from york

(PRE) Keyworded Peakmax (better than Ian Clifton's!), but
still recognizes old-style numerical input. New peakmax.doc,
peakmax.for

missed an LMB change.  (previous revisions were the wrong way round)

merged in previous AGWL changes

From Ian Clifton:
PEAKMAX has been keyworded as promised, I've also merged in some changes that
Inger asked me to do to our local copy of it. These are mainly to do with the
size of arrays, etc: it can find 8000 peaks instead of 800, though the default
is still 800 as before. You'll notice the following note in PEAKMAX's
documentation:
<<In the PDB/Diamond file on stream XYZOUT, the peaks are put out as
waters with a chain identifier of X until residue 999 is reached, when the
sequence continues from 1 and chain id Y. The official PDB format allows the
residue number to contain up to four digits, so this is technically
unnecessary.>>
I can't remember exactly why Inger thought 4-digit residue numbers are a problem
(there's no space between the first digit of that field and the preceding chain
ID), but I think Xplor was one of the culprits. Anyway, I guess it's a trivial
point.

copyright

add ccprcs

portability

from LMB (GF):
	(1)peaks sorted by abs.val. of height,
	(2)sorted list has height/sigma,
	(3)option for atom o/p to have occ=peak height,Btemp=height/sigma

fixed inline comments and inconsistent arg to sort1

Documentation for peakmax (peakmax.doc) updated to include option for
setting threshold as a multiple of rms value of map.  Change program
so that test for too small delta rho (interpolation) is done on a
multiple of rms value of map rather than an absolute value. Print
offending delta rho values. AGWL

Roughly correposnds to first `official' release of MTZ version for un*x
and VMS around 14--15/5/92.

Initial revision