Printer Friendly - TutorialCcp4Dls2016

Content:

Introduction
Setup
Example datasets and tutorials

Introduction

Here we will collect some information regarding use of our software (during tutorials or when using it on your own) that is particular to the setup during the DLS-CCP4 Data Collection and Structure Solution Workshop (December 13-20 2016).

We will try and keep this page here up-to-date as we get settled in during the course.

Setup

You might need to load a so-called module (every time you open a shell/terminal):

% module load ccp4-workshop
% module load sharp

NOTE: the current version of SHARP installed here is from September 2015, so not quite up-to-date. We are trying to fix that asap, but be prepared for some potential side-effects and issues.

There are different ways of running autoSHARP:

from within the CCP4i (not yet in CCP4i2) there is an autoSHARP task in

      Experimental Phasing
        --> Automated Search & Phasing
              --> autoSHARP

this has the advantage for users of CCP4i of presenting a familiar interface
however, looking at resulting maps and models requires a bit more work (one needs to download/save files first before loading them into Coot)
when loading MTZ files from SHARP/autoSHARP into Coot you need to select the correct columns for the type of map you want to look at (see here for details if the column naming is not already obvious
please ensure that you have configured CCP4i correctly to use your browser of choice (e.g. firefox) as so-called "hypertext" viewer

if you can type run_autoSHARP.sh -h in a terminal (and it comes back with the help message), then everything should be setup correctly for using the command-line interface to autoSHARP. You might want to have a printout of the autosharp_reference_card.pdf handy to help with usage.

be sure to be roughly familiar with Unix command line (terminal/shell). See e.g. some online tutorials:

this interface has the advantage of creating (at the end) little scripts to present results from various stages of structure solution easily within coot. You can run those using e.g.

          % coot --script 04_coot.scm

Example datasets and tutorials

There are several examples and tutorials available, e.g.

tutorials from the CCP4/Montevideo 2015 course: there are some really fast ones given which should easily finish in the short time we have available here

some suggested runs could be (to be faster - depending on the speed of the computing hardware at the course - just add the flag "-fast" to the command line: those jobs should still work fine with this):

SAD: 1o22.pir 1o22_peak.sca

          % run_autoSHARP.sh -seq 1o22.pir -ha "Se" \
                             -wvl 0.9778 peak -7 5 -sca 1o22_peak.sca \
                             -id autoSHARP_SAD | tee autoSHARP_SAD.lis

MAD: 3isy.pir 3isy_aimless_0.97934A.sca 3isy_aimless_0.91162A.sca

          % run_autoSHARP.sh -seq 3isy.pir -ha "Se" \
                             -wvl 0.97934 infl -11 3.3  -sca 3isy_aimless_0.97934A.sca \
                             -wvl 0.91162 hrem -1.8 3.3 -sca 3isy_aimless_0.91162A.sca \
                            -id autoSHARP_MAD | tee autoSHARP_MAD.lis

MIRAS: 3ZFT.pir 3ZFT_nat.sca 3ZFQ_Hg.sca 3ZFR_Ir.sca (this can run for a bit longer, but when running with the -fast switch it is possible the refined phases are not good enough for interpretation of density)

          % run_autoSHARP.sh -seq 3ZFT.pir \
                             -nat -sca 3ZFT_nat.sca \
                             -ha "Hg" -nsit 1 -wvl 1.54179 -sca 3ZFQ_Hg.sca \
                             -ha "Ir" -nsit 2 -wvl 1.54179 -sca 3ZFR_Ir.sca \
                             -id autoSHARP_MIRAS | tee autoSHARP_MIRAS.lis

more example datasets

or use your own data, especially if you have

a sequence file (see here for format description)
a reflection file (ideally a merged *.sca file, e.g. from running autoPROC)
information about heavy atom(s) (number, f'/f'' etc)