kennthshang / cherry_mags Goto Github PK

Hi, thank you for providing this usefull software! When I am using this software to explore the interaction between my phage contigs and my bacterial genomes, an issue regarding "KeyError" occurred (see a part of issue below). can you please help to address this issue?

------a part of issue---------
Creating blast database...
Running blastn...
Creating blast database...
Running blastn...
folder GCN_data exist... cleaning dictionary
Dictionary cleaned
Namespace(dataset='virus-host', gpus=0, learning_rate=0.001, mode='virus', model='pretrain', t=0.5, topk=1)
adj: (3578, 3578)
features: (3578, 256)
x : tensor([[0.8381, 0.7619, 0.4762, ..., 0.7238, 0.4381, 1.0000],
[0.9222, 0.9000, 0.4889, ..., 0.6444, 0.4444, 1.0000],
[0.0925, 0.0939, 0.1257, ..., 0.0773, 0.0939, 0.0925],
...,
[0.2126, 0.2518, 0.3834, ..., 0.2588, 0.2345, 0.2188],
[0.2297, 0.1968, 0.3956, ..., 0.2107, 0.2093, 0.2293],
[0.0521, 0.1075, 0.1238, ..., 0.0878, 0.0967, 0.0573]],
device='cuda:0')
sp: tensor([[0.5000, 0.5000, 0.0000, ..., 0.0000, 0.0000, 0.0000],
[0.5000, 0.5000, 0.0000, ..., 0.0000, 0.0000, 0.0000],
[0.0000, 0.0000, 0.0294, ..., 0.0000, 0.0000, 0.0000],
...,
[0.0000, 0.0000, 0.0000, ..., 0.2000, 0.0000, 0.0000],
[0.0000, 0.0000, 0.0000, ..., 0.0000, 0.3333, 0.0000],
[0.0000, 0.0000, 0.0000, ..., 0.0000, 0.0000, 0.1429]],
device='cuda:0')
input dim: 256
output dim: 256
num_features_nonzero: 0
Traceback (most recent call last):
File "run_Cherry.py", line 296, in
if crispr_pred[contig]:
KeyError: 'cherry_0_1506'
folder input exist... cleaning dictionary
folder pred exist... cleaning dictionary
folder Split_files exist... cleaning dictionary
Dictionary cleaned
folder tmp_pred exist... cleaning dictionary
Dictionary cleaned
folder train_phage/ exist... cleaning dictionary
Dictionary cleaned
folder node_feature exist... cleaning dictionary
Dictionary cleaned
GCN Error for file contig_{i}

The log file (nohup.out.txt) is attached for your information.
nohup.out.txt

The prokaryote.csv file has been edited according to the input bacterial genome, and a beginning part of this file is below:
Accession,Superkingdom,Phylum,Class,Order,Family,Genus,Species,Name
G1_1_2_bin_24,G1_1_2_bin_24,G1_1_2_bin_24,G1_1_2_bin_24,G1_1_2_bin_24,G1_1_2_bin_24,G1_1_2_bin_24,G1_1_2_bin_24,G1_1_2_bin_24
G1_1_2_bin_25,G1_1_2_bin_25,G1_1_2_bin_25,G1_1_2_bin_25,G1_1_2_bin_25,G1_1_2_bin_25,G1_1_2_bin_25,G1_1_2_bin_25,G1_1_2_bin_25
G1_1_2_bin_27,G1_1_2_bin_27,G1_1_2_bin_27,G1_1_2_bin_27,G1_1_2_bin_27,G1_1_2_bin_27,G1_1_2_bin_27,G1_1_2_bin_27,G1_1_2_bin_27

Hi, I run this pipeline on my own phage genomes and bacterial genomes to predict interactions between them. Based on the output, it seems that only one bacterial host is predicted for each phage genome. I wonder that how to predict all potential bacterial hosts of phage based on ones own phage and bacterial genomes using this pipeline? Thank you in advance!