Dear developer,
I truly want to train this RF-Score-VS with my own datasets to do further research, but I could find insructions in this Github page.
It would be nice if you could offer some information, thanks!

When running rfscore_vs with multiple CPUs with a large compound libraries, the system becomes irresponsible due to heavy load.

I noticed it writes temporary files to /tmp and well a large library implicates lots of writes ~227 Mb "compiledtrees_XXXXX.so" one for each core.

This process slows down system performance over time, even when /tmp is mounted on a SSDs. It is more pronounced when non-PRO SSDs are used after it's "fast-cache" fills ups.

Frequent "large" writes can degrade SSDs, as they have a limited number of write cycles to each NAMD cell.

I think it would be beneficial to write temporary files to RAM (/dev/shm/) as a way to speed things up systematically or at least give this option to the user.

I found there is an repository of the original author HongjianLi for RFScore, but few people use it.
However, many people use this rfscorevs_binary in GitHub. So is this repository able to represent the well-known RF-score mentioned in many articles?

Example of RFScore/RFScore-v3 been mentioned in reviews:

Hi, I am trying to test this scoring function, but the server is down and, as a result, binaries not accessible, nor the *.pickle files. Maybe they can be uploaded as a GitHub release for permanent storage? Thank you, Jaime.

Hi @mwojcikowski,

I am trying to use your tool to perform a rescoring of Vina's results, however I am not succeeding. I have ligands with different structures, but the tool generates identical scores values in all cases. Could you help me with a possible answer to the error I might be making?

Thanks

Dear developer,
When I use your binary RF-Score-VS to dock multiple complexes in PDBbind2019, they just get almost the same score (near 6), which is unbelievable terrible!

pdb_id,prot_name,rf_score_vs_v2,ref_affinity
1a1e,1a1e_pocket,6.00580229596,6.00
1a4k,1a4k_pocket,6.00794721751,8.00
1a4r,1a4r_pocket,6.20134056292,6.66
1a4w,1a4w_pocket,6.38927082738,5.92
1a9m,1a9m_pocket,6.1346691244,6.92
1a9q,1a9q_pocket,5.97711068812,6.17
1a28,1a28_pocket,7.25402944261,8.29
1a69,1a69_pocket,6.15717254042,5.30
1a94,1a94_pocket,6.76315462617,7.85
1a99,1a99_pocket,5.98602117631,5.70
1aaq,1aaq_pocket,6.08011277506,8.40
1add,1add_pocket,6.31263890627,6.74
1adl,1adl_pocket,5.97653251803,5.36
1ado,1ado_pocket,5.62528809888,6.00
1afk,1afk_pocket,6.07168576653,6.62
1afl,1afl_pocket,6.14445310932,6.28
1ai4,1ai4_pocket,6.06714064078,2.50
1ai5,1ai5_pocket,6.01511559545,3.72
1ai7,1ai7_pocket,5.97128016745,4.09
1aid,1aid_pocket,6.00623191904,4.82
1aj7,1aj7_pocket,6.04805370093,3.87
1ajn,1ajn_pocket,6.01279242565,2.63
1ajp,1ajp_pocket,6.12961282626,2.23

The following is part of my code:

    for prot_ in os.listdir(datasets_pro_dir):
        prot_name = prot_[:-4]
        pdb_id = prot_[:4]
        pocket_file = os.path.join(datasets_pro_dir, prot_)
        ligand_file = os.path.join(datasets_lig_dir, f'{pdb_id}_ligand.mol2')
        
        if pocket_file and ligand_file:
            command = f'./rf-score-vs --receptor {pocket_file} {ligand_file} -o csv --field "name" --field "RFScoreVS_v2"'
            output = subprocess.check_output(command, shell=True).decode('utf-8')
            rf_score = output.split('\n')[1].split(',')[1]
            writer.writerow({'pdb_id': pdb_id, 'prot_name': prot_name,'rf_score_vs_v2': rf_score})

Hello,

I am trying to run the binary using the latest version of oddt, anaconda, and relevant packages (rdkit, openbabel). I am receiving the following error:

File "/local_programs/rfscorevs/rf-score-vs", line 117, in
score_file = path_join(sys._MEIPASS if sys.frozen else dirname(abspath(file)), 'RFScoreVS_v2_vina.pickle')
AttributeError: module 'sys' has no attribute 'frozen'