* Postprocessing the ATG Prescreen

VFVS VFVS 2
#1

Hi all,

I am running Tutorial 2: VFVF2 2.0 ON SLURM and have reached * Postprocessing the ATG Prescreen | Tutorials (VF 2.0). The script vfvs_postprocess_atg-prescreen.sh was not included in the tutorial files, so I downloaded it from the Github tools/vfvs_postprocess_atg-prescreen.sh. In this script, vfvs_get_top_results.py assumes the use of AWS Batch, but since my environment uses SLURM, I commented out the relevant parts and granted execution permissions for Athena.

# if ctx['config']['batchsystem'] != 'awsbatch':
# print("In order to use this query script, AWS must be configured")
# print("TEST0")
# exit(1)

However, the following line in the output results in the awk error and Score is missing.

Completed 94 Bytes/94 Bytes (556 Bytes/s) with 1 file(s) remaining^Mdownload: s3://***/VFVS2-Tutorial1-Preconfigured-Slurm/tools/athena/***.csv to ./qvina02_rigid_receptor1.csv
Running query in AWS Athena

Waiting on createdb if needed (006e6db3-8ddd-43b0-b82a-1cd5906b75f3)
Waiting on dropping of old table (a4152c95-02dc-48f1-a457-59d5f691709e)
Waiting on create table (183fa670-4f2a-4753-bce0-a5dd83e6e64d)
Waiting on query (a232f6ed-79ce-4286-9ce1-2ac0d8172ee6)

STATUS: SUCCEEDED
Output location: s3://***/VFVS2-Tutorial1-Preconfigured-Slurm/tools/athena/***.csv
aws s3 cp {response[‘QueryExecution’][‘ResultConfiguration’][‘OutputLocation’]} …/output-files/qvina02_rigid_receptor1.csv
awk: cmd. line:1: fatal: division by zero attempted
Tranche,Class,Score
0,A,
0,B,
0,C,
0,D,
0,E,
0,F,
1,A,
1,B,

Looking at the output_files/qvina02_rigid_receptor1.csv ,

“ligand”,“collection_key”,“scenario”,“score_0”,“score_average”,“score_min”,“attr_smi”,“$path”
~
~
~

Although the column titles are being output, the values are missing, which seems to be causing the error mentioned above. Could you please explain why the values are missing and provide guidance on how to resolve this issue?

Additionally, what does the asterisk( * Postprocessing the ATG Prescreen) at the beginning of this section signify? Is it correct to assume that this section must be completed and cannot be skipped before proceeding to the next Preparing the ATG Primary Screen section?

Thank you in advance.

#2

Dear @y-komuro,

great that you are using VFVS 2.0 on Slurm! The current script is not compatible with Slurm (and would need further adjustments to work with it).

However, we have a version of the vfvs_get_top_results.py script for Slurm that is not yet pulled into the main repository. You can find it here: https://github.com/csecker/VFVS/blob/vfvs-2/tools/vfvs_get_top_results.py

You should be able to use it in combination with the vfvs_postprocess_atg-prescreen.sh script. Please let me know if you experience any problems.

All the best
Chris

#3

Dear @_Chris_Secker,

Thank you so much for your quick response.
When I used the script you provided, it worked fine! Then I ran this script again and confirmed that it ran successfully too in the final section (* Postprocessing the ATG Primary Screen | Tutorials (VF 2.0)). As a result, I was able to complete the ATG primary screen. Thank you again!

Just to be safe for other users, let me correct my post above.

  1. Instead of downloading vfvs_postprocess_atg-prescreen.sh from Git ( VFVS/tools/vfvs_postprocess_atg-prescreen.sh at vfvs-2 · VirtualFlow/VFVS · GitHub), I used gk1-tutorial/atg-prescreen/tools/vfvs_postprocess_atg-prescreen.sh.

  2. I ran the following commands as explained in this section ( Postprocessing the ATG Prescreen | Tutorials (VF 2.0))

$ aws s3 cp s3://<your AWS bucket access point ID>/Enamine_REAL_Space_2022q12.tranches.parquet ~/
$ aws s3 cp s3://<your AWS bucket access point ID>/Enamine_REAL_Space_2022q12.collections.parquet ~/
  1. Then the following command was carried out.
    $ ./vfvs_postprocess_atg-prescreen.sh 10000
#4

Sorry, the second item above was not necessary.

  1. Instead of downloading vfvs_postprocess_atg-prescreen.sh from Git ( VFVS/tools/vfvs_postprocess_atg-prescreen.sh at vfvs-2 · VirtualFlow/VFVS · GitHub), I used gk1-tutorial/atg-prescreen/tools/vfvs_postprocess_atg-prescreen.sh.

  2. Then the following command was carried out.
    $ ./vfvs_postprocess_atg-prescreen.sh 10000