Hi Zeyu, Thanks for testing this! I recompiled another executable that should use more of the CPU during the search. Can you please test it to see if this suits your needs? Naturally, if you set the number of threads too high it might degrade the overall performance of your machine during the Kojak search.
https://sourceforge.net/projects/sashimi/files/Trans-Proteomic%20Pipeline%20%28TPP%29/TPP%20v7.0%20%28Arafel%29%20rev%20X/Kojak.exe/download It would interesting to compare the performance of before and after this change. I am testing this on i9-13900H. Cheers, -David On Tue, May 28, 2024 at 9:21 PM Zeyu Wang <[email protected]> wrote: > Hi David, > > I tested it with the new version. Now it works well, but CPU usage problem > still exists. I change threads from 24, 20 to 16 and 8, but the TPP > searching still uses E-cores. Perhaps it doesn't support P-Cores with Intel > Hyper-Threading. I am just wondering what CPU you are using for Kojak.exe > running. > > Thanks again! Anyway, it can still run, though may take a bit longer time. > > Best, > Zeyu > > On Tuesday, May 28, 2024 at 11:17:55 PM UTC-4 David Shteynberg wrote: > >> Hello Zeyu, >> >> Sorry I don't know the answer to your question but if it helps I >> corrected the issue you identified in the previously distributed version >> compiled with gcc for the TPP distribution installer. >> I placed the gcc compiled version of Kojak.exe here: >> https://sourceforge.net/projects/sashimi/files/Trans-Proteomic%20Pipeline%20%28TPP%29/TPP%20v7.0%20%28Arafel%29%20rev%20X/Kojak.exe/download >> >> Perhaps this one exhibits a different behavior when running >> multi-threaded. In my tests the two version ran nearly at the same speed >> on my computer. >> >> Cheers, >> -David >> >> On Tue, May 28, 2024 at 7:27 PM Zeyu Wang <[email protected]> wrote: >> >>> Hi all, >>> >>> I notice another small issue (maybe this is my win11 problem?): when >>> running Kojak within TPP, I set the threads to 24 or 28. However, in the >>> task manager, I notice the main search was always running on exact >>> CPU16-31, which seem to be E-cores of this 13900K. Did anyone notice >>> similar problem? >>> >>> [image: Snipaste_2024-05-28_22-14-14.png] >>> [image: Snipaste_2024-05-28_22-15-57.png] >>> >>> Best, >>> Zeyu >>> >>> >>> >>> >>> >>> On Monday, May 27, 2024 at 11:12:09 AM UTC-4 Zeyu Wang wrote: >>> >>>> Hi David, >>>> >>>> Thanks! After replacing the current kojak.exe, It seems that now the >>>> kojak searching can continue without error message. >>>> >>>> Best, >>>> Zeyu >>>> >>>> On Friday, May 24, 2024 at 9:55:03 PM UTC-4 David Shteynberg wrote: >>>> >>>>> Hello again, >>>>> >>>>> After a bit of digging I found that this is likely caused by a bug in >>>>> the current TPP compiled and distributed version of Kojak that will have >>>>> to >>>>> be corrected for the next release. Meanwhile you should be able to >>>>> download and use the official Kojak build available here: >>>>> https://kojak-ms.systemsbiology.net/download.html >>>>> >>>>> >>>>> This version will require additional .net dependecies! >>>>> >>>>> Please replace the current kojak executable in c:/TPP/bin with the >>>>> executable from the zip file and try running kojak again. >>>>> >>>>> >>>>> Best of luck and keep us posted to your progress. >>>>> >>>>> -David >>>>> >>>>> On Fri, May 24, 2024, 1:07 PM Zeyu Wang <[email protected]> wrote: >>>>> >>>>>> Hi David, >>>>>> >>>>>> Thank you! Here are the two kojak parameter files I tested (the >>>>>> original and the customized). Link: Kojak_Troubleshooting >>>>>> <https://gtvault-my.sharepoint.com/:f:/g/personal/zwang3136_gatech_edu/EnxqUYYUZGNDvCEG8kB-vtsB3Q2GfNvcqaAygJ_0ttBEFA?e=Ub1nub> >>>>>> . >>>>>> >>>>>> Best, >>>>>> Zeyu >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> On Friday, May 24, 2024 at 3:47:38 PM UTC-4 David Shteynberg wrote: >>>>>> >>>>>> Hello Zeyu, >>>>>> >>>>>> Thanks for your interest in TPP and Kojak. Yes we maintain and use >>>>>> Kojak in the latest release. To help us troubleshoot this problem, would >>>>>> you mind sharing the parameter file used in your test? >>>>>> >>>>>> Thank you! >>>>>> -David >>>>>> >>>>>> On Fri, May 24, 2024, 12:28 AM Zeyu Wang <[email protected]> wrote: >>>>>> >>>>>> Hi, >>>>>> >>>>>> Are there anyone using Kojak in the TPP v7.0.0 ? Today I follow the >>>>>> instruction and try to run a quick test. But at the very beginning, it >>>>>> reports an error. >>>>>> >>>>>> "Kojak version 2.1.0, October 16 2023 Copyright Michael Hoopmann, >>>>>> Institute for Systems Biology Visit http://kojak-ms.org for full >>>>>> documentation. ****** Begin Kojak Analysis ****** Time at start: Fri May >>>>>> 24 >>>>>> 02:52:02 2024 Parameter file: c:/TPP/data/params/Kojak_2.conf 12 >>>>>> preconfigured crosslinkers. Reading FASTA database: >>>>>> c:/TPP/data/dbase/UP000005640_9606_20230222.fasta >>>>>> ... >>>>>> Total Proteins: 20594 >>>>>> Adding Kojak-generated decoys. New Total Proteins: 41188 >>>>>> 5658462 peptides to search (2785870 linkable). >>>>>> >>>>>> Fri May 24 02:52:45 2024 ERROR: >>>>>> ERROR: cannot open \c:\TPP\data\Test_Kojak\E_ZW_LPS_XL_F10.kojak.txt >>>>>> for output. Please ensure path and write permissions are correct." >>>>>> >>>>>> Is the error due to the output file path format? Or shall I adjust >>>>>> the read/write permission of the TPP? >>>>>> >>>>>> Best, >>>>>> Zeyu >>>>>> >>>>>> -- >>>>>> You received this message because you are subscribed to the Google >>>>>> Groups "spctools-discuss" group. >>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>> send an email to [email protected]. >>>>>> To view this discussion on the web visit >>>>>> https://groups.google.com/d/msgid/spctools-discuss/c1b41187-61c0-4dc2-ba42-9a997f09eb7en%40googlegroups.com >>>>>> <https://groups.google.com/d/msgid/spctools-discuss/c1b41187-61c0-4dc2-ba42-9a997f09eb7en%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>> . >>>>>> >>>>>> -- >>>>>> You received this message because you are subscribed to the Google >>>>>> Groups "spctools-discuss" group. >>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>> send an email to [email protected]. >>>>>> >>>>> To view this discussion on the web visit >>>>>> https://groups.google.com/d/msgid/spctools-discuss/9f93e2d2-c59b-4ed9-8813-4728ca2e4a16n%40googlegroups.com >>>>>> <https://groups.google.com/d/msgid/spctools-discuss/9f93e2d2-c59b-4ed9-8813-4728ca2e4a16n%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>> . >>>>>> >>>>> -- >>> You received this message because you are subscribed to the Google >>> Groups "spctools-discuss" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected]. >>> >> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/spctools-discuss/db56c834-5f2b-486f-a160-9b3bf9228ed2n%40googlegroups.com >>> <https://groups.google.com/d/msgid/spctools-discuss/db56c834-5f2b-486f-a160-9b3bf9228ed2n%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> >> -- > You received this message because you are subscribed to the Google Groups > "spctools-discuss" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/spctools-discuss/637fa2e7-7f6d-4811-9537-858bd6bb540bn%40googlegroups.com > <https://groups.google.com/d/msgid/spctools-discuss/637fa2e7-7f6d-4811-9537-858bd6bb540bn%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "spctools-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/CAGJJY%3D9%2Bv_5%2BEMMOAM6xJEGJyrsgmwOZrdohx%2B_H_hy3f23euw%40mail.gmail.com.
