Dear Ian, Thank you for this information. It worked! This useful feature is, as you said, not included in the documentation.
Also, I made the mistake to use the ccp4 gui to run Tlsanl. Not only would the gui insist that I had to provide a .tls file, when I left the space for the .tls file blank, it would pass an empty TLSIN parameter to the program, which unsurprisingly, Tlsanl failed to open resulting in the termination of the program. Cheers, Herman Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Ian Tickle Gesendet: Mittwoch, 11. Dezember 2013 12:03 An: CCP4BB@JISCMAIL.AC.UK Betreff: Re: [ccp4bb] Converting TLS to ANISO records Dear Herman TLSANL doesn't need a TLS file, it will read the PDB header if you don't give it one. Sorry if that's not explained in the documentation. It was a feature I added recently. Cheers -- Ian On 11 December 2013 10:59, <herman.schreu...@sanofi.com<mailto:herman.schreu...@sanofi.com>> wrote: Dear bulletin board, I recently submitted a coordinate file to the pdb that had been optimized with pdbredo, which used TLS refinement. All TLS parameters are in the header, but the pdb wants these to be converted to ANISO records. I did a google search and also looked into Tlsanl, but did not get a clear idea how to proceed. Tlsanl e.g. wants a .tls file, which I could not find amongst the pdbredo output files. Also I do not think I need a .tls file, since the information is in the pdb header. Also, at different times, different definitions of TLS parameters seemed to have been around. So what is the current practice to convert TSL parameters to ANISO records (Refmac 5.7.0029 was used)? Should I apply the overall anisotropic B factors to Fobs, or are these also integrated in the ANISO records, or should I leave these alone? Thank you for your help, Herman
