Yes, the Rfreeflag column should be of type I, see e.g. "Output:" section of
http://www.ccp4.ac.uk/dist/html/freerflag.html
And this is what ccp4i expects, see $CCP4/etc/types.def
But not all programs will check it.
In fact, the type changed from W to I in 1995, back in the days when free R was
still new and exciting.
Cheers
Martyn
-----Original Message-----
From: CCP4 bulletin board on behalf of iulek
Sent: Sun 5/18/2008 12:48 PM
To: [log in to unmask]
Subject: [ccp4bb] [Re: [ccp4bb] warp/arp 7.0.1 to really use the Free R flag] - Summary
Hi,
Concerning this point, it turned out that some scripts of mine (for
some historical good reasons) attribute "R" as column type for my
r-free-flag column. Changing it to "I" (which I think is more ccp4
standard) solved the problem. According to what I talked to Anastassis,
that must be a requirement from ccp4i. For several other programs there
was no problem (although I must recognize that I make a very limited use
of the ccp4 interface, ccp4i). Anyway, I will attach more to the
conventions, now.
Jorge
-------- Original Message --------
Subject: {Spam?} Re: [ccp4bb] warp/arp 7.0.1 to really use the Free R flag
Date: Tue, 29 Apr 2008 11:01:41 +0200
From: Anastassis Perrakis <[log in to unmask]>
To: [log in to unmask]
CC: [log in to unmask]
References: <[log in to unmask]>
Dear Jorge,
I was unable to reproduce the error you report with 7.0.1.
I selected an MTZ file, then chose 'use' FreeR flag and I run the job.
I get:
#CCP4I VERSION CCP4Interface 1.4.4.2
#CCP4I SCRIPT LOG arp_solvent
#CCP4I DATE 29 Apr 2008 10:56:28
#CCP4I USER Tassos
...
mtz labels taken: FP SIGFP FreeR_flag
...
FreeR will be used for monitoring and sigmaa calculations
...
Starting model: After refmac, R = 0.162 (Rfree = 0.170)
Could you please try to do the same with first building the standard
example (psp) and then adding waters?
It might be a problem with your mtz file.
Tassos
On Apr 28, 2008, at 15:22, [log in to unmask] wrote:
> Hi all,
>
> We recently installed arp/warp 7.0.1 under ccp4i (CCP4_6.0.2).
> The installation went fine. We just wanted to add water molecules to
> the current model. We toggle to "use" in the box
>
> "For refmac5 refinement do -use- the Free R flag"
>
> Also, we choose the due flag in the box
>
> Determine the free R label for reflections to be escluded from
> refinement -FreeR_flag-"
>
> When we ask the program to run, it does, but without using the
> Free R-flag; it says "FreeR will not be used" in the log file and
> the R-free is reported as 0.000. Curiously, the box where the free R
> label is becomes blank.
> We googled and searched archive lists (both warp and CCP4).
> What the problem might be ?
>
> Jorge
|