RitchieLabIGH / IRFinder

MIT License
13 stars 10 forks source link

IR-Finder with long readds #20

Open genotoul opened 1 year ago

genotoul commented 1 year ago

Can you precise in the doc what is changing with minimap2 when analysing long reads ?

CloXD commented 1 year ago

Hello, I included more details about the long mode ( https://github.com/RitchieLabIGH/IRFinder/wiki/Quantify-IR ). We didn't perform a full benchmark between STAR and minimap2 and the main changes between the short and long reads mode are in the IR quantification step. Let me know if what I added is enough and if you have some specific doubts. Cheers, Claudio

genotoul commented 1 year ago

Thank you Claudio,

Nothing specific to do for building the reference when using minimap2 ?

Christine

--

Christine GASPIN

INRAE, UPR 0875, MIA-T & PF GenoToul Bioinfo Chemin de Borde Rouge BP52 627 31326 Castanet Tolosan cedex @.*** Tel : 33 (0) 5 61 28 52 82 Fax : 33 (0) 5 61 28 53 35


De : Claudio Lorenzi @.***> Envoyé : vendredi 2 décembre 2022 16:24 À : RitchieLabIGH/IRFinder Cc : Christine Gaspin; Author Objet : Re: [RitchieLabIGH/IRFinder] IR-Finder with long readds (Issue #20)

Hello, I included more details about the long mode ( https://github.com/RitchieLabIGH/IRFinder/wiki/Quantify-IR ). We didn't perform a full benchmark between STAR and minimap2 and the main changes between the short and long reads mode are in the IR quantification step. Let me know if what I added is enough and if you have some specific doubts. Cheers, Claudio

- Reply to this email directly, view it on GitHubhttps://github.com/RitchieLabIGH/IRFinder/issues/20#issuecomment-1335414152, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A4QULT7YL3FP4VSU2D333GTWLIIBFANCNFSM6AAAAAASR76YGY. You are receiving this because you authored the thread.Message ID: @.***>

genotoul commented 1 year ago

Dear Claudio,

Following my previous mail, I mean, can you provide the way to build the reference and launch the aligment step with minimap2 as it is done for STAR ?

Here is my command:

IRFinder Long -r RET_INTRON_IRFinder -d NB05_irfinder /home/gaspin/work/STEFFEN/SAMPLES_2022/SUP/NB05_t5p3p.fastq

and the error message:

Argument error: -r RET_INTRON_IRFinder, Does not appear to be a valid IRFinder reference. Could not find RET_INTRON_IRFinder/IRFinder/ref-cover.bed

I suppose that the build reference step is to do ?

Christine

--

Christine GASPIN

INRAE, UPR 0875, MIA-T & PF GenoToul Bioinfo Chemin de Borde Rouge BP52 627 31326 Castanet Tolosan cedex @.*** Tel : 33 (0) 5 61 28 52 82 Fax : 33 (0) 5 61 28 53 35


De : Christine Gaspin Envoyé : vendredi 2 décembre 2022 18:08 À : RitchieLabIGH/IRFinder Objet : RE: [RitchieLabIGH/IRFinder] IR-Finder with long readds (Issue #20)

Thank you Claudio,

Nothing specific to do for building the reference when using minimap2 ?

Christine

--

Christine GASPIN

INRAE, UPR 0875, MIA-T & PF GenoToul Bioinfo Chemin de Borde Rouge BP52 627 31326 Castanet Tolosan cedex @.*** Tel : 33 (0) 5 61 28 52 82 Fax : 33 (0) 5 61 28 53 35


De : Claudio Lorenzi @.***> Envoyé : vendredi 2 décembre 2022 16:24 À : RitchieLabIGH/IRFinder Cc : Christine Gaspin; Author Objet : Re: [RitchieLabIGH/IRFinder] IR-Finder with long readds (Issue #20)

Hello, I included more details about the long mode ( https://github.com/RitchieLabIGH/IRFinder/wiki/Quantify-IR ). We didn't perform a full benchmark between STAR and minimap2 and the main changes between the short and long reads mode are in the IR quantification step. Let me know if what I added is enough and if you have some specific doubts. Cheers, Claudio

- Reply to this email directly, view it on GitHubhttps://github.com/RitchieLabIGH/IRFinder/issues/20#issuecomment-1335414152, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A4QULT7YL3FP4VSU2D333GTWLIIBFANCNFSM6AAAAAASR76YGY. You are receiving this because you authored the thread.Message ID: @.***>

CloXD commented 1 year ago

Hello Christine, sorry for the late answer. The reference is the same as for the short read mode: minimap2 won't need the STAR reference built in that step, but all the other files generated are required for the following ones. Cheers Claudio

genotoul commented 1 year ago

Dear Claudio,

Thank you for the answer. I have the fasta and gtf files in the directory for the reference but it doesn't work with the given command line (see my previous mail.

What are the needed input files ?

Perhaps you can precise that in the doc.

Christine

--

Christine GASPIN

INRAE, UPR 0875, MIA-T & PF GenoToul Bioinfo Chemin de Borde Rouge BP52 627 31326 Castanet Tolosan cedex @.*** Tel : 33 (0) 5 61 28 52 82 Fax : 33 (0) 5 61 28 53 35


De : Claudio Lorenzi @.***> Envoyé : lundi 5 décembre 2022 09:15 À : RitchieLabIGH/IRFinder Cc : Christine Gaspin; Author Objet : Re: [RitchieLabIGH/IRFinder] IR-Finder with long readds (Issue #20)

Hello Christine, sorry for the late answer. The reference is the same as for the short read mode: minimap2 won't need the STAR reference built in that step, but all the other files generated are required for the following ones. Cheers Claudio

- Reply to this email directly, view it on GitHubhttps://github.com/RitchieLabIGH/IRFinder/issues/20#issuecomment-1336931243, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A4QULT66AIYOTQAUPDR72QDWLWQCBANCNFSM6AAAAAASR76YGY. You are receiving this because you authored the thread.Message ID: @.***>

CloXD commented 1 year ago

Hello Christine, sure, the reference is here: https://github.com/RitchieLabIGH/IRFinder/wiki/Build-Reference

So, assume you have your custom genome reference and annotation in ./custom_genome.fasta and ./custom_annotation.gtf and you want to create your reference in ./IRFinder_REF

mkdir ./IRFinder_REF/
cp ./custom_genome.fasta ./IRFinder_REF/genome.fa  ### or you can use mv or ln to move or create a link
cp ./custom_annotation.gtf ./IRFinder_REF/transcripts.gtf
IRFinder BuildRefProcess  -r ./IRFinder_REF/
### -e -b and -R are for the SpikeIn, blacklist and region of interest, not mandatory

Cheers, Claudio