search

shendurelab / MIPGEN

One stop MIP design and analysis
Other
22 stars 19 forks source link

using known isoform and oligos #29

Closed JulieHachey closed 1 year ago

JulieHachey commented 7 years ago

Hi,

Would it be possible to generate mips against a known sequence without having to run it against the reference genome and skipping the BWA part? I would like to use RNA isoforms and short oligos.

Would it be possible to input our own Mip backbone?

Thanks!!

augustboyle commented 7 years ago

Hello,

BWA is used to determine the specificity of the probe so it does have to be run, BUT you are welcome to use any reference you like to design the MIPs. You can include all the oligos and isoforms you want to design to and build your own BWA index. Then you would need to figure out the coordinates of the region you want to have tiled.

The only caveats are that if you are capturing cDNA from an RNA-seq library, I would recommend using the whole transcriptome of the organism, and for scoring purposes, some flanking sequence is required beyond what you want to tile exactly. Otherwise, you could design non-specific probes that will yield lots of side products / the scoring will not work.

Good luck!

Evan

On October 3, 2017 at 5:10:55 AM, JgitHhub (notifications@github.com) wrote:

Hi,

Would it be possible to generate mips against a known sequence without having to run it against the reference genome and skipping the BWA part? I would like to use RNA isoforms and short oligos.

Would it be possible to input our own Mip backbone?

Thanks!!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/shendurelab/MIPGEN/issues/29, or mute the thread https://github.com/notifications/unsubscribe-auth/AF01p6lP6EpUxLzhF8YuE9QTzkq9HFxgks5soiRPgaJpZM4PsAsQ .

JulieHachey commented 7 years ago

Thank you for the information.

Would it be possible to import a fasta and have mips generated against the sequences provide instead of using the reference to find a sequence?

augustboyle commented 7 years ago

Hi,

It is straightforward to run bwa index on any fasta you like to generate a new reference, so that shouldn’t be a problem! The variable is called bwa_genome_index, but it doesn’t have to be a genome fasta. Just toss it in.

The only other thing is that gap filling MIPs perform poorly on DNA shorter than about 1000 bp, so if you have very short libraries it might be quite inefficient.

Evan

On October 3, 2017 at 10:52:46 AM, JgitHhub (notifications@github.com) wrote:

Thank you for the information.

Would it be possible to import a fasta and have mips generated against the sequences provide instead of using the reference to find a sequence?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/shendurelab/MIPGEN/issues/29#issuecomment-333925481, or mute the thread https://github.com/notifications/unsubscribe-auth/AF01p5rkA374WEEJOJUtRK_1MqH_5rN1ks5sonRtgaJpZM4PsAsQ .