Why aren’t you using SHELX? Diana ************************************************** Diana R. Tomchick Professor Departments of Biophysics and Biochemistry UT Southwestern Medical Center 5323 Harry Hines Blvd. Rm. ND10.214A Dallas, TX 75390-8816 [email protected]mailto:[email protected] (214) 645-6383 (phone) (214) 645-6353 (fax) On May 1, 2019, at 3:22 PM, Whitley, Matthew J mailto:[email protected]> wrote: Hello all, We have collected electron diffraction data to approximately 1.8 Å on a 'small molecule' of roughly 1660 Da. The space group (P 43 21 2) and unit cell parameters determined during data reduction match the known, published values, and our data are roughly 80% complete in all resolution shells from 17 - 1.8 Å. We would like to solve this structure by molecular replacement in Phaser, but are having a bit of difficulty doing so. Our problems all seem to be related to specifying the composition of the target molecule for Phaser. Via the phenix GUI, we can only specify the target composition as either protein or nucleic acid, when in fact the target is neither. The chemical composition of the target is C72 H85 N19 O18 S5. Is there a way to feed this chemical composition to Phaser and to get it to understand that the target is neither protein nor nucleic acid? Furthermore, according to our reading of the Phaser documentation, specifying "protein" leads to an assumed solvent content of 50%, which is likely to be a substantial overestimation for this small molecule crystal. We have also run into the problem of Phaser complaining that the target molecule will not fit into the given unit cell, which we think is due to the assumption of 50% solvent content. We would be thankful for whatever advice you may have on using Phaser with non-protein, non-nucleic acid molecules. Matthew --- Matthew J. Whitley, Ph.D. Research Instructor W. Furey Lab Department of Pharmacology & Chemical Biology University of Pittsburgh School of Medicine _______________________________________________ phenixbb mailing list [email protected]mailto:[email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]mailto:[email protected] ________________________________ UT Southwestern Medical Center The future of medicine, today.

Dear Matthew, I'm really pleased to hear that you're trying this out! I've played a bit with using Phaser on test cases of large "small molecules", and it seems to work well, but I've had a hard time convincing people who work on such things that it should be tested on unsolved structures. I'll be particularly interested to hear how you get on with electron diffraction data. Probably because this feature is not used that much, we hadn't really noticed that the Phenix GUI doesn't allow the specification of non-macromolecular components in the composition. So I'm afraid that to get this to work, you'll have to use the scripting interface. If you haven't used that, there are sample scripts on our website, e.g. the first one here: http://www.phaser.cimr.cam.ac.uk/index.php/MR_using_keyword_input http://www.phaser.cimr.cam.ac.uk/index.php/MR_using_keyword_input. To make that work in your case, you would have to substitute the COMPOSITION command with a separate command for each element type, as explained here: http://www.phaser.cimr.cam.ac.uk/index.php/Keywords#COMPOSITION http://www.phaser.cimr.cam.ac.uk/index.php/Keywords#COMPOSITION. So for instance, you could have a command like COMPOSITION ATOM C NUMBER 72 for the 72 C atoms expected in your asymmetric unit. Note that what you specify in different COMPOSITION commands just gets added up. There are electron form factors for electron diffraction, so you just need to add the command FORMFACTORS ELECTRON to your script to account for that (though it may not make much difference — it would be interesting to know!). Phaser's happy with any of the 230 space groups in any setting, because it just takes the symmetry from the MTZ file. The only aspect of symmetry that might be a problem is that if you have a racemic mixture and the two hands are both present in the asymmetric unit (i.e. without being generated by an inversion operator) you have to provide separate models. I hope this will all work! However, we've encountered some design issues with the handling of composition in the current version of Phaser, which we've corrected by redesigning the way this is handled in the upcoming rewritten version of Phaser. So if you're still running into trouble getting Phaser to accept the size of what you're looking for, please get directly in touch and send us a log file so we can sort out a way (even if only temporary) to make this work for you! Best wishes, Randy Read

On 1 May 2019, at 21:22, Whitley, Matthew J wrote:

Hello all,

We have collected electron diffraction data to approximately 1.8 Å on a 'small molecule' of roughly 1660 Da. The space group (P 43 21 2) and unit cell parameters determined during data reduction match the known, published values, and our data are roughly 80% complete in all resolution shells from 17 - 1.8 Å. We would like to solve this structure by molecular replacement in Phaser, but are having a bit of difficulty doing so.

Our problems all seem to be related to specifying the composition of the target molecule for Phaser. Via the phenix GUI, we can only specify the target composition as either protein or nucleic acid, when in fact the target is neither. The chemical composition of the target is C72 H85 N19 O18 S5. Is there a way to feed this chemical composition to Phaser and to get it to understand that the target is neither protein nor nucleic acid? Furthermore, according to our reading of the Phaser documentation, specifying "protein" leads to an assumed solvent content of 50%, which is likely to be a substantial overestimation for this small molecule crystal. We have also run into the problem of Phaser complaining that the target molecule will not fit into the given unit cell, which we think is due to the assumption of 50% solvent content.

We would be thankful for whatever advice you may have on using Phaser with non-protein, non-nucleic acid molecules.

Matthew

--- Matthew J. Whitley, Ph.D. Research Instructor W. Furey Lab Department of Pharmacology & Chemical Biology University of Pittsburgh School of Medicine _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]

------ Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: + 44 1223 336500 The Keith Peters Building Fax: + 44 1223 336827 Hills Road E-mail: [email protected] Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk