search

nanoporetech / ont_fast5_api

Oxford Nanopore Technologies fast5 API software
Other
144 stars 28 forks source link

demultiplexing generates additional - folder #68

Closed Fatihlrcfs closed 2 years ago

Fatihlrcfs commented 2 years ago

Hi Dear, I am working on one of the barcoded fast5 files demultiplexing and ı am using this < demux_fast5 --input fast5_skip/ --save_path demultiplexed_reads/ --summary_file sequencing_summary_FAP54362_e8e82a0a.txt > sequencing summary was provided by ONT Minknown. During demultiplexing additional from the barcode folders (-) folder is generated. Also, all demultiplex fast5 is placed into that folder and my barcode folder is empty. I already base called this data and ı did not face the same problem while ı was working and generating fastq files. now ı am working on another analysis program and ı needs single fast5 so ı am doing demultiplexing. how can ı corretly classify my demultiplex fast5 files from the skipped files. thanks

fbrennen commented 2 years ago

Hi @Fatihlrcfs -- demux_fast5 sorts reads that have already been demultiplexed and had a barcode classification assigned to them. The fast5_skip folder contains reads that MinKNOW "skipped" and did not basecall or demultiplex -- this is why they all get put into that - folder (meaning there is no demultiplexing result for them). If you want to demultiplex the reads from the skip folder you will need to use guppy.

Fatihlrcfs commented 2 years ago

Hi @fbrennen sorry for my disturbance but ı am quite new to this area and ı don't know how to deal with this problem. ı need single fast5 files for Deeepsignal and Tombo analysis. I read the guppy and did not find how to generate single fast5 files from the fast_skip using guppy. if it is possible could you explain me more basic ? manny thanks. Sincerely

fbrennen commented 2 years ago

Hi @Fatihlrcfs -- not a problem. The multi_to_single_fast5 script will convert multi-read fast5 files to single-read ones that are suitable for use by tombo (and presumably Deepsignal, though I have never used that). If they come from the fast5_skip folder then you will likely need to generate some basecalling results for those reads using Guppy.

I would recommend dropping by the Nanopore Community to ask general questions -- there are far more people there, and they might be able to recommend some good quickstarts or tutorials. https://nanoporetech.com/community