Quantify antibiotic resistance protein families on bacterial isolate genomes

[kuangd...@gmail.com]

Hi,

I know shortBRED is a good tool to quantify relative abundance of the interested proteins in metagenome or metatranscriptome. I have hundreds of Salmonella genomes which have different antibiotic resistance phenotypes by serotypes. I am planning to generate a figure to profile RPKM values for antibiotic resistance protein families in different strains, like Kaminski J did in the human gut microbiome in his paper “High-specificity targeted functional profiling in microbial communities with ShortBRED”. I am wondering if it also makes sense to quantify antibiotic resistance protein families in bacterial isolate genomes, instead of in microbial communities.

Thanks in advance,

Dai

[Violetta Florova]

I have the same question, actually.

I tried to run whole bacterial genome - I've got error. 

But it's definitely should be good idea.

[Jim Kaminski]

Hello,

My apologies for the delay in response. I've set aside some time this weekend to go through unanswered questions on the message board and will get back to both of you soon.

Best,

Jim

--
You received this message because you are subscribed to the Google Groups "shortbred-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to shortbred-users+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jim Kaminski]

Hello Dai, Hello Violetta,

Yes - you can use ShortBRED to quantify antibiotic resistance families in isolate genomes. I recommend using fasta files of ORF's in the isolate genomes - it's what we did in the ShortBRED paper.

You can find instructions on how to do this at https://bitbucket.org/biobakery/shortbred/wiki/Home in the section: "Method 2: Profile a set of ORFs from isolate genome (an "annotated" genome) using ShortBRED-Quantify and a set of markers" .

Best,

Jim

[Violetta Florova]

Dear Jim,

Thank you for your explanation again.

I run agains  ShortBRED_ARDB_Evaluation_markers - it works, but I didn't have any hits.

When i run against  ShortBRED_ABR_101bp_markers.faa

python /PATH_to_directory/shortbred_quantify_test.py --genome genome_AZY.fna

--markers ShortBRED_ABR_101bp_markers.faa --tmp bacterialgenome

--usearch /PATH_to_directory/src/usearch --maxhits 0 --maxrejects 0 --results results_ORF_AZY.txt

I've got this error 

/projects/academic/mjbuck/Buck_Lab_Members/violetta/new_project/src/usearch --usearch_local ShortBRED_ABR_101bp_markers.faa --db bacterialgenome/genome_AZY.fna.udb --id 0.95 --userout bacterialgenome/annotated_genomefull_results.tab --userfields target+query+id+alnlen+mism+opens+qlo+qhi+tlo+thi+evalue+bits+ql+tl+qs+ts --maxaccepts 0.0 --maxrejects 0.0 --threads 1

---Fatal error---

Must specify -strand plus or both with nt db

Traceback (most recent call last):

  File "/projects/academic/mjbuck/Buck_Lab_Members/violetta/new_project/shortbred_quantify_test.py", line 411, in <module>

    iAccepts=args.iMaxHits, iRejects=args.iMaxRejects,strUSEARCH=args.strUSEARCH )

  File "/projects/academic/mjbuck/Buck_Lab_Members/violetta/new_project/src/quantify_functions.py", line 299, in RunUSEARCHGenome

    "--maxrejects",str(iRejects),"--threads", str(iThreads)])

  File "/util/academic/biopython/1.70/anaconda-5.0.1/lib/python2.7/subprocess.py", line 186, in check_call

    raise CalledProcessError(retcode, cmd)

subprocess.CalledProcessError: Command '['/projects/academic/mjbuck/Buck_Lab_Members/violetta/new_project/src/usearch', '--usearch_local', 'ShortBRED_ABR_101bp_markers.faa', '--db', 'bacterialgenome/genome_AZY.fna.udb', '--id', '0.95', '--userout', 'bacterialgenome/annotated_genomefull_results.tab', '--userfields', 'target+query+id+alnlen+mism+opens+qlo+qhi+tlo+thi+evalue+bits+ql+tl+qs+ts', '--maxaccepts', '0.0', '--maxrejects', '0.0', '--threads', '1']' returned non-zero exit status 1

How can I "specify strand"?

Thanks for your help,

To unsubscribe from this group and stop receiving emails from it, send an email to shortbred-use...@googlegroups.com.