tseemann / legsta

🍗⭐ In silico Legionella pneumophila Sequence Based Typing

legsta

In silico Legionella pneumophila Sequence Based Typing (SBT)

Background

SBT stands for sequence-based typing. The purpose of the Legionella pneumophila SBT scheme is to provide a rapid and easily comparable method for the epidemiological typing of clinical and environmental isolates of Legionella pneumophila in outbreak investigations.

Install

Conda

conda install -c conda-forge -c bioconda -c defaults legsta

Homebrew

brew install brewsci/bio/legsta

Github

cd $HOME
git clone https://github.com/tseemann/legsta.git
cd $HOME/legsta/test
../bin/legsta *.fna *.gbk

Input

The any2fasta tool is used to convert input files to FASTA for feeding to isPcr. It can accept FASTA, Genbank, EMBL, GFF, and many other formats. The files may also be compressed with gzip, bzip2 or zip.

Output

Output is a TSV file (or CSV if --csv is used). Alleles with no in silico product are denoted - and novel alleles listed using ?.

% cd legsta/test
% ../bin/legsta NC_006368.fna NC_018140.fna CR628336.1.gbk.gz missing_flaA.fna FJBS01000000.fna.bz2


FILE  		       SBT     flaA    pilE    asd     mip     mompS   proA    neuA
NC_006368.fna          1       1       4       3       1       1       1       1
NC_018140.fna          734     2       6       17      1       1       8       11
CR628336.1.gbk.gz      1       1       4       3       1       1       1       1
missing_flaA.fna       -       -       14      16      25      7       13      206
FJBS01000000.fna.bz2   -       3       10      1       3       14      9       ?

Options

Option	Description
--quiet	do not print any informational messages to stderr
--csv	comma-separated output instead of tab-separated
--noheader	don't print table header to output (ie. FILE SBT flaA ...)
--version	print legsta X.Y.Z version to stdout and exit

Dependencies

• Perl >= 5.26
• any2fasta
• isPcr - Linux | Mac | Source

Issues

Submit questions or issues to our Issue Tracker

Authors

• Torsten Seemann
• Anders Goncalves Da Silva
• Andrew Buultjens
• Jason Kwong

Licence

GPLv3

Acknowledgements

• Natalie Groves for providing the latest sequences and profiles from the PHE database

References

• Public Health England - L.pneumophila SBT Home Page
• Gaia et al (2005)
• Ratzow et al (2007)

legsta's Issues

Unwanted output saved to redirected file

At the moment, if we run:

legsta myfasta.fa > legsta_res.txt

The first line of the file turns out to be:

-minPerfect=N - Minimum size of perfect match at 3' end of primer (default 15)

Turns out, on line 30 of legsta we have this:

system("$ISPCR 2>&1 | grep -- minPerfect")==0 or err("Can not run $ISPCR");

Perhaps it should be:

system("$ISPCR 2>&1 | grep -- minPerfect > /dev/null")==0 or err("Can not run $ISPCR");

Use any2fasta to support lots of input types

flaA allele 11 not detected

Hi, thanks for making this tool, it's very useful! I am running it on 70 genomes which I also have Sanger SBT on to compare. For most of the genomes, legsta finds the same SBT, but it does not seem to find flaA allele 11. 25 of the genomes (SBTs 154 and 574) should have flaA 11, but legsta outputs "-" for this allele. Other flaA alleles are found in the other genomes. When I run BLASTn on the assemblies using the flaA 11 sequence from https://github.com/tseemann/legsta/blob/master/db/flaA.tfa, I find flaA_11 with 100% nucleotide identity and sequence coverage.

Also, for 12 of those 25 genomes, the mompS-gene was called incorrectly (as allele 7 in stead of 15). The mompS-allele is correctly called in all the other genomes I have looked at.

Any idea what is going on?

Thanks,
Marit

pilE 10 allele is not found

It seems there are some mismatches in the PCR primers that cause isPCR to fail for that particular allele.

Add --quiet mode

Just print the table

SBT from amplicons

Hi,
Great tool for implementing L. pneumophila SBT! Thanks.
I'm trying to call SBT using nanopore sequenced multiplex PCR amplicons. However, I get "no product" results for some of the alleles. Sequences for the alleles are in the input file and are a perfect match for an allele in the database. I tested an assembled genome of the same SBT strain and get expected results.
Can I use amplicons with Legsta? Do the input sequences need to be longer?

Cheers!

ERROR: Could not find 'any2fasta'. Please install it and ensure it is in the PATH.

Hi Seemann,
I got the is after I cloned the repository and rand this command

../bin/legsta *.fna *.gbk (while in the test directory)

ERROR: Could not find 'any2fasta'. Please install it and ensure it is in the PATH.
which path?
Any ideas?
Thanks1
TJ

ST update

Hi,

is it possible to get an update of https://github.com/tseemann/legsta/blob/master/db/legionella.txt ?
I tried to find one at http://bioinformatics.phe.org.uk/legionella/legionella_sbt, but the web site is no more available.

Thanks
Eric

Databases are out of date

I think they are behind a password.

neuA.fta headers

Hi @tseemann,

we found a discrepancy in the fta file for the neuA loci. Two header formats exist:

neuA_xx
neuAh_xx

was this intentional? may this affect the results (i.e. unassigned allele for this loci, etc)?

thanks,
Yair

Add --label option

Instead of filename, print --label.
Only works for 1 sample.

Perhaps --label could be a regexp to capture a part of the filename

Hi Torsen

Hi Torsen, i am wondering if the script works fine with concatenation of the 7 sequences in one fasta file instead of a whole genome sequence. We have only data from first generation sequencing. Actually, i tried it and it doesn't works.

Thanks
Eric

Handle empty FASTA file better

./bin/legsta /dev/null
FILE    SBT     flaA    pilE    asd     mip     mompS   proA    neuA
Running: isPcr /dev/null /home/tseemann/git/legsta/bin/../db/ispcr.tab stdout -minPerfect=6 -tileSize=6 -maxSize=1200 -stepSize=5 -out=fa
Program error: trying to allocate 0 bytes in needLargeMem (limit: 1073741824)
isPcr: memalloc.c:91: needLargeMem: Assertion `FALSE' failed.
/dev/null       -       -       -       -       -       -       -

might need to tweak isPcr settings a bit

First sequence I tested it on was /mnt/seq/GENBANK/bacteria/Legionella/pneumophila/str._Lens/
Had a mismatch in the primer at position 6 for mip allele - worked once I changed the isPcr settings to -minPerfect=5 instead of -minPerfect=6

Could test some of the strains in this table: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3877988/table/T3/

No result returned when finding a sequence with isPcr but it does not match exactly with anything in the database

Running legsta on an isolate with what looks like to be a new mompS allele returns a null result, as the allele is not in the database.

It would be nice to return a partial match, rather than a null result.

Add flag to save/print isPcr output

It would be good to have the option to save the output from isPcr.

I can get it by asking for --debug twice.