zheminzhou / etoki Goto Github PK

It would be very helpful to be able to pip3 install EnSuit.

Good morning,
I hope it is relevant here.
Could EToKi be used to search for an AA motif in the EnteroBase db?

Thank you in advance

El

Hi, could you clarify in the documentation which things I can put into --path? Like if I have the entire spades package downloaded, do I just run it with --path spades.py=$(which spades.py)? For blast, do I need to consider multiple executables like makeblastdb and blastn?

Thank you!

Hello,
I installed EToKi and tried the example as showed in the README file but I got the following error.

./EToKi.py MLSTdb -i examples/Escherichia.Achtman.alleles.fasta -r examples/Escherichia.Achtman.references.fasta -d examples/Escherichia.Achtman.convert.tab
2020-07-29 20:45:04.997470 Exemplar sequences in ./NS_2085r4ap/clsFna.clust.exemplar
2020-07-29 20:45:04.997569 Clusters in ./NS_2085r4ap/clsFna.clust.tab
2020-07-29 20:45:05.030599 Run BLASTn starts
2020-07-29 20:45:05.384751 Run BLASTn finishes. Got 971 alignments
2020-07-29 20:45:05.384888 Run diamond starts
2020-07-29 20:45:06.047658 Run diamond finishes. Got 966 alignments
2020-07-29 20:45:06.375628 removed 0 paralogous sites.
2020-07-29 20:45:06.375682 obtained 5530 alleles and 39 references alleles
2020-07-29 20:45:06.378881 A file of reference alleles has been generated: examples/Escherichia.Achtman.references.fasta
Traceback (most recent call last):
File "EToKi.py", line 47, in
etoki()
File "EToKi.py", line 41, in etoki
eval(arg.cmd)(sys.argv[2:])
File "/mnt/data/disk1/biotools/EToKi/modules/MLSTdb.py", line 164, in MLSTdb
conversion[0].append(get_md5(allele['value']))
TypeError: string indices must be integers

I installed in a conda env with python 3.6.
Some ideas how to fix it?

It would be much nicer if all the compenents were "sub commands" of a single tool.

Having understandable names would be even better.

Thanks 👍

http://cab.spbu.ru/ website, where the "SPAdes-3.13.0-Linux.tar.gz" file is downloaded from, is down.

With phylo.py, the core genome output doesn't match expectations. Assuming I'm correct in thinking that the .matrix is the right output.

With my alignment and a threshold of 0.95, 4,800,000/5,000,000 sites should be retained. Instead, 4,781,411 are output.

python EToKi/EToKi.py -h
Doesn't handle Help when no module is given

Traceback (most recent call last):
  File "EToKi/EToKi.py", line 55, in <module>
    etoki()
  File "EToKi/EToKi.py", line 31, in etoki
    exec('from modules.{0} import {0}'.format(sys.argv[1]))
  File "<string>", line 1
    from modules.-h import -h

requests is required by package, but not mentioned in readme.

I am taking some notes on how I ran cgMLST, and I hope you can add documentation for it.

Create database: this took a very long time

# Downloaded the cgMLST scheme from enterobase FTP into Salmonella.cgMLSTv2.enterobase (undocumented)
\ls -f1 Salmonella.cgMLSTv2.enterobase/*.fasta | \
  grep -v cgMLST_v2_ref.fasta `# ignore already-established reference file` | \
  xargs seqtk seq -l 0 `# cat out all the fasta contents and two-line fasta format` | \
  perl -lane '
    # get the id with '>' and the seq on the next line since it is in a two-line fasta format
    $id=$F[0]; 
    $seq=<>; 
    chomp($seq); 
    # I don't think this will matter but just avoid any infinite loops by quitting if we see the same sequence
    my %seen; 
    if($seen{$id}++){print STDERR "Already seen $id. Done."; last;} 

    # Avoid deflines that might be problematic
    if($id =~ /[^_>0-9a-zA-Z]/){
      print STDERR "Skipping ".$id; 
      next;
    } 
    print "$id\n$seq";
  ' > enterobase.filtered.fasta

Hi @zheminzhou ,

am trying to recreate the SNP analysis that would be performed on EnteroBase, given a directory of assemblies.

Would it be possible to add a note to the README about what programs correspond to the steps listed in the EnteroBase docs?

For what I can tell, it seems like:

• refMasker ~ RecHMM
• refMapper ~ align
• refMapper_matrix ~ RecFilter
• matrix_phylogeny ~ phylo

... but if that's the case, I'm not sure where the SNP matrix required by RecHMM would come from; the docs describe refMasker as identifying recombination regions from a reference genome, while RecHMM identifies regions from a SNP matrix.

Any help would be appreciated!

Thanks,

~Nick

Hi, I am wondering if it will work if I download vsearch and rename it to usearch in my PATH. I want to package EToKi into a container but the usearch license is prohibitive.