Hi Yuan,

I used TLS+ADP refinement to get the final structure model, after that I used phenix.model_vs_data to get the statistics for table1. However there is no B factor statistics in the output.

- most of the necessary statistics is in phenix.refine .log file and in PDB file header. - phenix.model_vs_data reports pretty extended ADP statistics (for macromolecule, solvent, ligands, etc..). For example, it has something like this: ADP (min,max,mean): all (136 atoms): 4.4 44.1 10.9 side chains (48 atoms): 4.8 23.0 9.5 main chains (64 atoms): 4.4 9.8 6.4 macromolecule (112 atoms): 4.4 23.0 7.7 ligands (1 atom): 6.6 6.6 6.6 solvent (23 atoms): 8.8 44.1 26.8 and if you don't see this then please make sure you are using a recent PHENIX version (I added this at least 6 months ago). What are the numbers you are looking for ?

Can I get the overall B factor by just simply averaging the B factors of each residue in the output text? Then how can I get the B factor value of the sidechains and that of the mainchain?

You don't need to do all this since the existing tools do it all for you. Pavel.