search

andrewjpage / krocus

Predict MLST directly from uncorrected long reads
GNU General Public License v3.0
25 stars 6 forks source link

Krokus miscall with E coli MLST #5

Closed gingerdave269 closed 6 years ago

gingerdave269 commented 6 years ago

Hi Andrew, We had two samples which we sequenced via Nanopore which we know are STEC ST11 (12,12,8,12,15,2,2) using our E. coli MLST database, we used krocus (default parameters) with the same MLST database to sequence type them and originally got both samples coming back as an SLV of ST - ST7806 (12,1068,8,12,15,2,2).

I played around with the parameters a bit such as --kmer 15, --margin 500 and managed to get one of the samples samples as ST11 however, the other is still an SLV of ST11 (ST7806).

Our ONT sequences are on the NCBI SRA under SRR7477813 and SRR7477814 if you wish to access them.

We were wondering if you could help us?

Thanks! David.

andrewjpage commented 6 years ago

Hi David, Great to hear your using the software. I've fixed the underlying issue so it should work out of the box on your data. There was a 9 base repeat in the middle of the gene which caused it to get a bit confused, but its working now (I only tested it on SRR7477813 as I'm on holidays atm). Let me know how you get on. Andrew

gingerdave269 commented 6 years ago

Hi Andrew, Thanks for this, I really appreciate it!

fabienneaf commented 4 years ago

Hi Andrew, So far Krocus was working very good for our bacterial isolates after Oxford-nanopore sequencing. :) We have been running assays for S. aureus and K. pneumoniae sequence reads. However, this week we have run for the first time the database 1 for E. coli. Although our sequencing looks very good (we did the quality control and there is an average of 100X of coverage among the isolates) we did not get results when using Krocus for it. All our 6 E.coli isolates resulted in "ND" ST. When I tried to run the database 2 for E. coli, It ran very good again and I got perfect STs for all the isolates. Is there some issue about the database 1? Any clue? Thank you! Fabienne

andrewjpage commented 4 years ago

Hi, Thanks for the feedback. In the first instance try downloading the latest version just in case theres something with the DB. krocus_database_downloader -o databases/Escherichia_coli_1 -s 'Escherichia coli#1' I've run the Escherichia coli#1 DB over a range of nanopored ecoli and the performance was disappointing. Using a different method (galru, coming soon) I doubled the number of STs recovered, however in that dataset there was a wide range of coverage so its not generalisable. Andrew

On Fri, 28 Feb 2020 at 10:13, fabienneaf notifications@github.com wrote:

Hi Andrew, So far Krocus was working very good for our bacterial isolates after Oxford-nanopore sequencing. :) We have been running assays for S. aureus and K. pneumoniae sequence reads. However, this week we have run for the first time the database 1 for E. coli. Although our sequencing looks very good (we did the quality control and there is an average of 100X of coverage among the isolates) we did not get results when using Krocus for it. All our 6 E.coli isolates resulted in "ND" ST. When I tried to run the database 2 for E. coli, It ran very good again and I got perfect STs for all the isolates. Is there some issue about the database 1? Any clue? Thank you! Fabienne

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/andrewjpage/krocus/issues/5?email_source=notifications&email_token=AAAF4V6XEWIL5LB3XDUUWUDRFDPT7A5CNFSM4FPBH6H2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOENH75ZQ#issuecomment-592445158, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAF4V2JQZ4C5UFSOIGCDTDRFDPT7ANCNFSM4FPBH6HQ .

fabienneaf commented 4 years ago

Thank you very much, Andrew. I`ll try to download the latest version of the database then.