search

Novartis / EQP-cluster

Unix-based RNA-seq quantification pipeline
Apache License 2.0
10 stars 3 forks source link

Using longer reads #3

Open mjafin opened 8 years ago

mjafin commented 8 years ago

Hi there, Many thanks for releasing this code to the community. Regarding the statement "due to the way the custom junction Fasta file is created EQP-cluster works best with reads of length 101bp or less" is there a way to improve the pipeline to work well with longer reads? We have quite often 150bp reads that we'd like to use

sven0schuierer commented 8 years ago

Hi Miika,

You can use EQP-cluster to process reads of length 150bp as well. You will lose alignments against junctions that are a) not part of the junctions in the transcripts and b) where the overlap to either side of the junction is less than 50bp. This is not ideal but the gene level quantifications will be minimally affected. I am currently looking into ways to make other read lengths available.

Best regards,

Sven

From: Miika Ahdesmaki [mailto:notifications@github.com] Sent: Tuesday, June 21, 2016 4:47 PM To: Novartis/EQP-cluster Subject: [Novartis/EQP-cluster] Using longer reads (#3)

Hi there, Many thanks for releasing this code to the community. Regarding the statement "due to the way the custom junction Fasta file is created EQP-cluster works best with reads of length 101bp or less" is there a way to improve the pipeline to work well with longer reads? We have quite often 150bp reads that we'd like to use

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/Novartis/EQP-cluster/issues/3, or mute the threadhttps://github.com/notifications/unsubscribe/AL1XKwALGzYLDmMUzzeTPaEAqHBlDs3wks5qN_lZgaJpZM4I60Xa.

mjafin commented 8 years ago

Hi Sven, Thanks for the prompt reply. I think our main focus would be on exon skipping where the event would not be described by standard isoforms. Do you think we could pull this off with the current pipeline using 150bp reads?

sven0schuierer commented 8 years ago

Hi Miika,

For this purpose using the 101bp junction data base for the alignment would be limiting. I can create the 150bp instance but getting it out of Novartis will be a bit of a problem. I’ll have to enquire how this can be done.

Best regards,

Sven

From: Miika Ahdesmaki [mailto:notifications@github.com] Sent: Wednesday, June 22, 2016 10:32 AM To: Novartis/EQP-cluster Cc: Schuierer, Sven; Comment Subject: Re: [Novartis/EQP-cluster] Using longer reads (#3)

Hi Sven, Thanks for the prompt reply. I think our main focus would be on exon skipping where the event would not be described by standard isoforms. Do you think we could pull this off with the current pipeline using 150bp reads?

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/Novartis/EQP-cluster/issues/3#issuecomment-227677689, or mute the threadhttps://github.com/notifications/unsubscribe/AL1XK5Sc6SzUDLuSqXTnfRdJHoSNh_Tbks5qOPLugaJpZM4I60Xa.

mjafin commented 8 years ago

OK, no worries, let us know how it goes