ABACUS stuck at " Creating combined table from 'interact-ALL.prot.xml'"

[Lindsay]

Hello,

I've been trying to run ABACUS a few times today and each time it freezes around 40% done at "Creating combined table from 'interact-ALL.prot.xml'" step.

I ran from the cmd line (shown below). Computer has 4.00 GB RAM, 32-bit OS, Windows 7 Professional. FASTA is 48 MB. Any idea what is jamming up so that I can run ABACUS to completion? Thank you very much!  Lindsay

[Lindsay]

[Lindsay]

Let me just add as a postscript that the previous runs were all done with "C[161]" entered in the "Req AA Mod" box (I am not quite sure how to use this field correctly, but since I used iodoacetamide I assume that I want to treat Carbamidomethyl-C as a req mod - please let me know if this is entered correctly).

On a whim I left the "Req AA Mod" field blank and ran to the same point, but with a different error message (below). I appreciate any advice you can give me!

Best,

 

Lindsay 

[Lindsay]

Also, I want to run with 6 prot/pep.xml pairs. If I delete most of them and run ABACUS with only one prot/pep.xml pair, it runs to completion in less than a minute.

[Damian]

Hi Lindsay.

I'm sorry you are having problems with Abacus.
The error message suggests that you are running out of memory for the data you are trying to process.
When you run ABACUS check the box that says "Keep DB". This will slow down the program a bit but you should be able to run the program to completion.
Please try this and let me know if it helps

Damian

---------------------------------------------------------------
Damian Fermin, Ph.D.
damian...@gmail.com
http://damianfermin.com
734-395-7664

> <Untitled2.png>

[Damian]

To use the Req. AA Mod field place the modification you want in the text field.
So for your example (C[161]) you would type in: +C[161]

---------------------------------------------------------------
Damian Fermin, Ph.D.
damian...@gmail.com
http://damianfermin.com
734-395-7664



On Oct 29, 2012, at 8:57 PM, Lindsay wrote:

> <Untitled2.png>

[Damian]

If you want to run 6 prot/pep.xml pairs you will need a total of 7 XML files:
6 pep.xml files
7 prot.xml files (one of them is the COMBINED file).

If you created the combined file and in the process created a corresponding pep.xml file for it, delete the combined.pep.xml file. It will definitely cause ABACUS to run out of memory.
How did you create the COMBINED file?

Damian

---------------------------------------------------------------
Damian Fermin, Ph.D.
damian...@gmail.com
http://damianfermin.com
734-395-7664

[dfermin]

Based upon your earlier posts. You should delete the interact-ALL.pep.xml file if it exists.

This should allow ABACUS to run to completion.

Hope this helps.
Let me know if you make any progress.

Damian

On Monday, October 29, 2012 9:04:32 PM UTC-4, dfermin wrote:

If you want to run 6 prot/pep.xml pairs you will need a total of 7 XML files:
6 pep.xml files
7 prot.xml files (one of them is the COMBINED file).

If you created the combined file and in the process created a corresponding pep.xml file for it, delete the combined.pep.xml file. It will definitely cause ABACUS to run out of memory.
How did you create the COMBINED file?

Damian

[Lindsay]

Hi Damian,

 

Thank you very much for your quick response! In a nutshell, I can run to completion now with the DB option checked, but only if I leave the "Req AA Mod" field blank -- see below for more details:

 

1. Sorry I wasn't clear about the pep/prot.xml pairs. In fact I had everything as you wrote:

"If you want to run 6 prot/pep.xml pairs you will need a total of 7 XML files:
6 pep.xml files

7 prot.xml files (one of them is the COMBINED file)."...with NO interact-ALL.pep.xml file.

I only meant to say that when I ran ABACUS with fewer input files, I was able to run to completion, but when I increased the input files, it stalled.

 

2. "When you run ABACUS check the box that says "Keep DB". This will slow down the program a bit but you should be able to run the program to completion.

Please try this and let me know if it helps"

Yes, with this setting I was able to run to completion, however, only in the case that I leave the "Req. AA Mod field" blank.

 

3. If I type in: +C[161] into "Req. AA Mod field" (and run as above) I get the following error: (attached). In my XTandem parameters I definitely had Carbamidomethyl-C as a fixed modification, so I am not sure why nothing in my combined file would meet the input parameters.

I don't remember previous versions of ABACUS having the "Req AA Mod" field and I am wondering how valid the output is if I leave the field blank?

 

4. How did you create the COMBINED file?

I created this by moving all the individual pep.xml files to one folder and then using xinteract to combine them into a combined.pep.xml file. Then I ran ProteinProphet on the combined.pep.xml file to get a combined prot.xml file.

 

Thanks again for all of your help! It is much appreciated.

 

Lindsay

[dfermin]

So we've made progress. Good!
Remove the C[161] from the Req AA Mod box.
That field is meant for variable modifications not for fixed ones. (I will ding myself for not making that clear in the manual).

Let me know how things go from here.

Damian

[Lindsay]

Hi Damian,

 

Thanks again for your reply!

 

If I remove +C[161] from Req AA Mod box, I can run to completion. However, I want to make sure that I am using the Req AA Mod field correctly:

 

"That field is meant for variable modifications not for fixed ones."

 

The variable modifications I used in my XTandem searches were:

15.994915@M,8.014199@K,10.008269@R

First, I wonder what is the difference between the + and - prefix in the examples shown (+S[167], -M[147])? Since I believe oxidized methionine has a larger mass than methionine and phosphorylated serine has a larger mass than serine I wonder why the M has a - prefix and the S has a + prefix.

So should my variable modifications be entered into Req AA Mod Box as:

 

-M[147], +K[136], +R[166]                   ?

 

(the - sign on M[147] I am basing on the example, the + sign on K and R I am just guessing)

Thanks!

 

Lindsay

[dfermin]

The sign (+ or -) refers to inclusion or exclusion of peptides with the given modification.

Assume you have these 2 peptides:
ELVIST[181]YPE
ELVIST[181]YP[97]E

The first peptides has a +80 on threonine and the second one has the T[181] plus a made up modification on Proline for this example.

Let's say you want to consider any and all proteins that have peptides with a T[181] modification but not a P[97] modification.

In the Req. AA. Mod text field you would enter: +T[181]; -P[97]

This would tell Abacus to keep the first peptide but exclude the second one.

The square bracket syntax is meant to be consistent with how modifications are reported in the TPP xml files. So M+16 appears as M[147] in the XML files.

If you don't put anything in the Req. AA. Mod text field, all peptides will be considered so long as they meet your probability threshold.

In your case, if you want to ensure that you are only counting peptides with M+16, K+8, and R+10 then enter:
+M[147]; +K[136]; +R[166]

into the Req AA Mods  text field.

You can see the affect of this by telling Abacus to produce peptide-level results, you should only see peptides with these 3 modifications in them.

Hope this helps,

Damian





On Tuesday, October 30, 2012 1:55:49 PM UTC-4, Lindsay wrote:

Hi Damian,

 

Thanks again for your reply!

 

If I remove +C[161] from Req AA Mod box, I can run to completion. However, I want to make sure that I am using the Req AA Mod field correctly:

 

"That field is meant for variable modifications not for fixed ones."

 

The variable modifications I used in my XTandem searches were:

15.9...@M,8.014199@K,10.008269@R

[Lindsay]

Hi Damian,

 

Thanks for your clear explanation, I think I understand that option much better now. In this case I will just leave the field blank. I am looking at my output now and it seems reasonable. I want to run SAINT now so it seems that I should use total spectral counts (not adjusted) to be more conservative in my assessment.

 

Thanks again for your help!

 

Lindsay