search

bjornwallner / DockQ

DockQ is a single continuous quality measure for Protein, Nucleic Acids and Small Molecule Docking Models
MIT License
212 stars 49 forks source link

Script fixnumbering.pl does not output model.pdb.fixed #17

Closed JasperGeissler closed 5 months ago

JasperGeissler commented 1 year ago

Hi,

Thank you for developing this useful software. When I attempt to run the script to align the protein lengths of the separate chains of the predicted model, I am met with the error that needle is not installed. However, when I run the script without any input, it says needle is in fact installed. How do I circumvent this?

Thanks in advance.

nemo8130 commented 1 year ago

To check if needle is installed and running, type 'needle' and enter from any terminal. If it is installed and working, you should be prompted the following:

@.**:~$ needle* Needleman-Wunsch global alignment of two sequences Input sequence:

Best, Sankar

On Fri, Mar 10, 2023 at 5:42 PM JasperGeissler @.***> wrote:

Hi,

Thank you for developing this useful software. When I attempt to run the script to align the protein lengths of the separate chains of the predicted model, I am met with the error that needle is not installed. However, when I run the script without any input, it says needle is in fact installed. How do I circumvent this?

Thanks in advance.

— Reply to this email directly, view it on GitHub https://github.com/bjornwallner/DockQ/issues/17, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACMQMFJSQQG75HN5J57ZHZDW3MLDVANCNFSM6AAAAAAVWMIYTE . You are receiving this because you are subscribed to this thread.Message ID: @.***>

JasperGeissler commented 1 year ago

Thank you for your reply.

I have tried this already, and verified that needle is installed and available from the command line. Using the command 'which needle' I have also verified that needle is available from the command line. Invoking the script fix_numbering.pl without any inputs provides an extra confirmation by returning "needle found to be installed in the path". This message changes to "needle NOT found in the path" upon removal of emboss, so needle is most definitely there.

My issue pertains to the script somehow not being able to detect needle when I invoke the script with the predicted and native .pdb files as per your documentation (by using scripts/fix_numbering.pl model.pdb native.pdb with DockQ as my working directory). I was wondering if you have a solution for this unexpected behaviour.

Edit: I installed emboss through conda using the command conda install -c bioconda emboss. Is there a specific version of emboss required?

nemo8130 commented 1 year ago

It would be best if Prof. Wallner replies to this - as he long maintains it. I still use an older GitHub version, which runs fine for me. And, one has to use fix_numbering.pl for making the two input files continuous in sequence.

@.***:~/Desktop/my_softwares/DockQ$ ./DockQ.exe EXAMPLE_PDBS/model_fixed.pdb EXAMPLE_PDBS/native_fixed.pdb

I haven't needed to use it lately.

Best, Sankar

On Fri, Mar 10, 2023 at 8:28 PM JasperGeissler @.***> wrote:

Thank you for your reply.

I have tried this already, and verified that needle is installed and available from the command line. Using the command 'which needle' I have also verified that needle is available from the command line. Invoking the script fix_numbering.pl without any inputs provides an extra confirmation by returning "needle found to be installed in the path". This message changes to "needle NOT found in the path" upon removal of emboss, so needle is most definitely there.

My issue pertains to the script somehow not being able to detect needle when I invoke the script with the predicted and native .pdb files as per your documentation (by using scripts/fix_numbering.pl model.pdb native.pdb with DockQ as my working directory). I was wondering if you have a solution for this unexpected behaviour.

— Reply to this email directly, view it on GitHub https://github.com/bjornwallner/DockQ/issues/17#issuecomment-1463924843, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACMQMFPUG2SQW7CYYYEKLFDW3M6P5ANCNFSM6AAAAAAVWMIYTE . You are receiving this because you commented.Message ID: @.***>

JasperGeissler commented 1 year ago

I have e-mailed professor Wallner as well earlier, and he unfortunately hasn't had opportunity to reply yet.

Have you and prof. Wallner maintained contact? Do you think it would be possible then for you to reach out to him as well? I hope that is not too tall an order.

Thank you for the help.

bjornwallner commented 1 year ago

Strange. A quick fix would be to remove the check for location of needle in the fix numbering script and hard code the full path when calling needle

We are working on an update in which does not require any additional scripts. It wil be out in the next couple of weeks.

Cheers Björn

On 10 Mar 2023, at 16:40, JasperGeissler @.***> wrote:



I have e-mailed professor Wallner as well earlier, and he unfortunately hasn't had opportunity to reply yet.

Have you and prof. Wallner maintained contact? Do you think it would be possible then for you to reach out to him as well? I hope that is not too tall an order.

Thank you for the help.

— Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fbjornwallner%2FDockQ%2Fissues%2F17%23issuecomment-1463983733&data=05%7C01%7Cbjorn.wallner%40liu.se%7C1aa4f5a6072942414e9908db217dbcf7%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638140596142095734%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=egj5STG4HMIrskHM32W%2FBihWwXe3D2FS%2FwvkSC87EMM%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FABAG3YDBL6JA7W3XCZK3O23W3NDNVANCNFSM6AAAAAAVWMIYTE&data=05%7C01%7Cbjorn.wallner%40liu.se%7C1aa4f5a6072942414e9908db217dbcf7%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638140596142095734%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=8NFnFFz4e1EwwywPip9o96UqoaOEU3yzNsPIggvuaKY%3D&reserved=0. You are receiving this because you are subscribed to this thread.Message ID: @.***>

JasperGeissler commented 1 year ago

Hi, the issue appears to have been solved. There was a bug on my end, the script works as intended. Thank you for the troubleshooting and the development of this useful software.

JasperGeissler commented 1 year ago

Hi, I now have a different issue (sorry for the inconvenience). It appears the script fixnumbering.pl does not output the desired fixed pdb model required for quality control. Has this issue arisen before?

nemo8130 commented 1 year ago

👍

On Mon, Mar 13, 2023, 2:50 PM JasperGeissler @.***> wrote:

Hi, the issue appears to have been solved. There was a bug on my end. Thank you for the troubleshooting and the development of this useful software.

— Reply to this email directly, view it on GitHub https://github.com/bjornwallner/DockQ/issues/17#issuecomment-1465779442, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACMQMFLPSEKCCJ6ZYB5DMW3W33RF7ANCNFSM6AAAAAAVWMIYTE . You are receiving this because you commented.Message ID: @.***>

rocke2020 commented 1 year ago

Strange. A quick fix would be to remove the check for location of needle in the fix numbering script and hard code the full path when calling needle We are working on an update in which does not require any additional scripts. It wil be out in the next couple of weeks. Cheers Björn On 10 Mar 2023, at 16:40, JasperGeissler @.> wrote:  I have e-mailed professor Wallner as well earlier, and he unfortunately hasn't had opportunity to reply yet. Have you and prof. Wallner maintained contact? Do you think it would be possible then for you to reach out to him as well? I hope that is not too tall an order. Thank you for the help. — Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fbjornwallner%2FDockQ%2Fissues%2F17%23issuecomment-1463983733&data=05%7C01%7Cbjorn.wallner%40liu.se%7C1aa4f5a6072942414e9908db217dbcf7%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638140596142095734%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=egj5STG4HMIrskHM32W%2FBihWwXe3D2FS%2FwvkSC87EMM%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FABAG3YDBL6JA7W3XCZK3O23W3NDNVANCNFSM6AAAAAAVWMIYTE&data=05%7C01%7Cbjorn.wallner%40liu.se%7C1aa4f5a6072942414e9908db217dbcf7%7C913f18ec7f264c5fa816784fe9a58edd%7C0%7C0%7C638140596142095734%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=8NFnFFz4e1EwwywPip9o96UqoaOEU3yzNsPIggvuaKY%3D&reserved=0. You are receiving this because you are subscribed to this thread.Message ID: @.>

I also have problem to install the 2013/2012 year EMBOSS-6 and cannot find needle at all. Sincerely waiting for a new version without dependent on EMBOSS. thanks!!

rocke2020 commented 1 year ago

@nemo8130 "We are working on an update in which does not require any additional scripts. It wil be out in the next couple of weeks." I see the Dock Q from many CAPRI papers, an excellent contribution for us. It is quite hard to me to write a CAPRI metrics, especially i-rmsd(it need pre align and then calculcate interface area), fnat, and now I do need a continuous docking quality score. If you can update Dock Q without these additon scripts, it will be very good helpful to us. thanks!!!