*** For details on how to be removed from this list visit the *** *** CCP4 home page http://www.ccp4.ac.uk ***
Areaimol preserves the value of Icode. When the input pdb file is read in, the value of Icode is stored in a character string, along with various other parameters such as residue number. When data is written to the output pdb file, the value of Icode is extracted from the string and written to the output pdb file. You can verify this by taking a small pdb file such as toxd.pdb in the CCP4 examples. Manually renumber some of the residues (e.g. change 57,58,59 to 56A,56B,57) and run areaimol. You will see the insertion code letters in both the output pdb file and the output written to the screen. As you point out, areaimol loses the altloc information that was present in the input pdb file. It also loses the original occupancy values. When it produces the output pdb file, the altloc field is set to blank and the occupancy field set to 1.0. However the atoms with fractional occupancy are not lost - all the other fields still appear in the output. As far as the solvent area calculation is concerned, areaimol appears to treat atoms with fractional occupancy on an equal footing with atoms whose occupancy is 1.0. I think the warnings you got with 1a34 occurred for a similar reason as 1ej6 - Hg HETATOMs had been changed to ATOMs. Norman -----Original Message----- From: Dan Bolser [mailto:[EMAIL PROTECTED] Sent: 29 November 2006 14:06 To: Stein, ND (Norman) Cc: [email protected] Subject: Re: [ccp4bb]: Some questions about Areaimol? Stein, ND (Norman) wrote: > Hi Dan > > If I run areaimol on 1ej6.vdb (rather than 1ej6_full.vdb) I get the following > warning message, which seems more appropriate: > > > *** Warning(s): point code unit function > *** 1 16512 1 MMDB_F_Atom > *** file : 1ej6.vdb > *** warning: unknown form factor encountered > atom type found and not recognised = > ZN H2000 > > $TEXT:Warning: $$ comment $$ > WARNING: Unknown atom type: adding to list with default radius 1.8 A > $$ > > This is caused by the last atom in the vdb file. For some reason, someone has > taken the original 1ej6 pdb file and changed the last record (the zinc atom) > from a HETAM to an ATOM record in creating the vdb file. It is these zinc > atoms that are giving the warning messages you see. > > Norman Stein > CCP4 > Thanks for the information. It seems that all the HETATM lines have been changed to ATOM lines in ViperDB. Can I bug you again about the Icode / AltLoc fields and AREAIMOL? How does it handle these attributes with regard to input / output? I guess all AltLocs are dropped when the atoms with fractional occupancy are dropped, but what about Icodes? Thanks for any help, Dan. > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of > Dan Bolser > Sent: 14 November 2006 13:20 > To: Dan Bolser > Cc: CCP4 > Subject: Re: [ccp4bb]: Some questions about Areaimol? > > > *** For details on how to be removed from this list visit the *** > *** CCP4 home page http://www.ccp4.ac.uk *** > > > Dan Bolser wrote: > >> *** For details on how to be removed from this list visit the *** >> *** CCP4 home page http://www.ccp4.ac.uk *** >> >> >> I am calculating the ASA of some virus structures downloaded from >> VIPERdb. >> >> I get the following error when processing some of the 'full capsid' >> structures (after tweaking MAXNET as appropriate)... >> >> *** Warning(s): point code unit function >> *** 1 16576 1 MMDB_F_Atom >> *** file : DBd/ViperDb/structures/1ej6_full.vdb.gz >> *** warning: atom serial number does not match position >> *** position unavailable, serial number 34487 >> *** warning: unknown form factor encountered >> *** position unavailable, serial number 34487 >> >> >> *** Warning(s): point code unit function >> *** 1 16576 1 MMDB_F_Atom >> *** file : DBd/ViperDb/structures/1ej6_full.vdb.gz >> *** warning: atom serial number does not match position >> *** position unavailable, serial number 34487 >> *** warning: unknown form factor encountered >> *** position unavailable, serial number 34487 >> >> >> ... (and many many more). >> >> I tried Googling, but I couldn't find the cause of this error. >> Following the errors (many thousands), the ASA appears to be >> calculated correctly (for at least some of the files that I have). >> Actually the above structure in particular is problematic in other >> ways... >> >> ERROR: Dimension of parameter MAXNET is too small >> Currently it is : 200000000 >> Needs to be at least: 1368998631 >> >> >> Which I think is impossibly large - Not sure though. >> >> >> My question is, what is this warning? Is it safe to ignore? Is it >> related to issues with Areaimol or rather the structure from VIPERdb? >> >> I had thought this warning was related to 'ALT LOC' codes in the >> files, for example, >> >> *** Warning(s): point code unit function >> *** 1 16576 1 MMDB_F_Atom >> *** file : DBd/ViperDb/structures/1a34_full.vdb.gz >> *** warning: atom serial number does not match position >> *** position unavailable, serial number 668 >> *** warning: unknown form factor encountered >> *** position unavailable, serial number 668 >> atom type found and not recognised = >> THR A 13 >> $TEXT:Warning: $$ comment $$ >> WARNING: Unknown atom type: adding to list with default radius 1.8 A >> $$ >> >> >> *** Warning(s): point code unit function >> *** 1 16576 1 MMDB_F_Atom >> *** file : DBd/ViperDb/structures/1a34_full.vdb.gz >> *** warning: atom serial number does not match position >> *** position unavailable, serial number 669 >> *** warning: unknown form factor encountered >> *** position unavailable, serial number 669 >> >> >> *** Warning(s): point code unit function >> *** 1 16576 1 MMDB_F_Atom >> *** file : DBd/ViperDb/structures/1a34_full.vdb.gz >> *** warning: atom serial number does not match position >> *** position unavailable, serial number 670 >> *** warning: unknown form factor encountered >> *** position unavailable, serial number 670 >> >> >> .... >> >> The above warnings appear to be co-localized with 'alternate location >> codes' in the PDB. How does Areaimol handle AltLocs (and Icodes)? >> >> >> Cheers, >> >> Dan. >> >> >> >> >> > > Does Areaimol simply ignore the ALTLOC codes? > > >
