search

MikeAxtell / ShortStack

ShortStack: Comprehensive annotation and quantification of small RNA genes
MIT License
88 stars 29 forks source link

Cluster analysis using ShortStack #70

Closed arazit closed 6 years ago

arazit commented 6 years ago

Dear Dr. Axtell, I have used ShortStack 3.8.3 to predict small RNA clusters. I noticed that the more small RNAs I analyze I get less clusters. This happened also when I use constant mincov and not relative mincov (rpm). I use foldsize = 1000. Do you have any idea why? Thanks, Tzahi

MikeAxtell commented 6 years ago

Hi Tzahi,

Yes, this is expected. Bascially, with deeper coverage but the same mincov, clusters grow larger and adjacent clusters merge together ... if there are two hotspots, they will be found as two separate clusters with low coverage data. But if you sequence deeper, it often is the case that there are a few reads found in between hotspots, which cause the two initial clusters to be merged into one.

Best, Mike

On Wed, Jan 24, 2018 at 2:06 AM, arazit notifications@github.com wrote:

Dear Dr. Axtell, I have used ShortStack 3.8.3 to predict small RNA clusters. I noticed that the more small RNAs I analyze I get less clusters. This happened also when I use constant mincov and not relative mincov (rpm). I use foldsize = 1000. Do you have any idea why? Thanks, Tzahi

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/70, or mute the thread https://github.com/notifications/unsubscribe-auth/AGiXiR6ZqPjtLI4gLq9VC59LuRtXFm0Jks5tNtZjgaJpZM4RqzvD .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

arazit commented 6 years ago

Hi Mike Thanks a lot. Can I use deep coverage and still for example detect Mirnases? Tzahi

-------- Original message -------- From: Mike Axtell notifications@github.com Date: 1/24/18 18:00 (GMT+02:00) To: MikeAxtell/ShortStack ShortStack@noreply.github.com Cc: Tzahi Arazi tarazi@volcani.agri.gov.il, Author author@noreply.github.com Subject: Re: [MikeAxtell/ShortStack] Cluster analysis using ShortStack (#70)

Hi Tzahi,

Yes, this is expected. Bascially, with deeper coverage but the same mincov, clusters grow larger and adjacent clusters merge together ... if there are two hotspots, they will be found as two separate clusters with low coverage data. But if you sequence deeper, it often is the case that there are a few reads found in between hotspots, which cause the two initial clusters to be merged into one.

Best, Mike

On Wed, Jan 24, 2018 at 2:06 AM, arazit notifications@github.com wrote:

Dear Dr. Axtell, I have used ShortStack 3.8.3 to predict small RNA clusters. I noticed that the more small RNAs I analyze I get less clusters. This happened also when I use constant mincov and not relative mincov (rpm). I use foldsize = 1000. Do you have any idea why? Thanks, Tzahi

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/70, or mute the thread https://github.com/notifications/unsubscribe-auth/AGiXiR6ZqPjtLI4gLq9VC59LuRtXFm0Jks5tNtZjgaJpZM4RqzvD .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/MikeAxtell/ShortStack/issues/70#issuecomment-360179306, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AiGgjcAVV2cLktcdnZuyIE8CTm4w5cm2ks5tN1IHgaJpZM4RqzvD.

=ee-email-cont> This mail scanned by ARO IT mail secure system powered by PineApp Cybonet --ef.-1.dfa535fd3fb3d1ea08030b8ac3dcb74e.ef--

MikeAxtell commented 6 years ago

Definitely. You may just need to adjust the mincov if you feel the clustering is too aggressive. Sometimes each dataset takes a little tuning.

On Wed, Jan 24, 2018 at 11:19 AM, arazit notifications@github.com wrote:

Hi Mike Thanks a lot. Can I use deep coverage and still for example detect Mirnases? Tzahi

-------- Original message -------- From: Mike Axtell notifications@github.com Date: 1/24/18 18:00 (GMT+02:00) To: MikeAxtell/ShortStack ShortStack@noreply.github.com Cc: Tzahi Arazi tarazi@volcani.agri.gov.il, Author < author@noreply.github.com> Subject: Re: [MikeAxtell/ShortStack] Cluster analysis using ShortStack (#70)

Hi Tzahi,

Yes, this is expected. Bascially, with deeper coverage but the same mincov, clusters grow larger and adjacent clusters merge together ... if there are two hotspots, they will be found as two separate clusters with low coverage data. But if you sequence deeper, it often is the case that there are a few reads found in between hotspots, which cause the two initial clusters to be merged into one.

Best, Mike

On Wed, Jan 24, 2018 at 2:06 AM, arazit notifications@github.com wrote:

Dear Dr. Axtell, I have used ShortStack 3.8.3 to predict small RNA clusters. I noticed that the more small RNAs I analyze I get less clusters. This happened also when I use constant mincov and not relative mincov (rpm). I use foldsize = 1000. Do you have any idea why? Thanks, Tzahi

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/70, or mute the thread https://github.com/notifications/unsubscribe-auth/ AGiXiR6ZqPjtLI4gLq9VC59LuRtXFm0Jks5tNtZjgaJpZM4RqzvD .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/ MikeAxtell/ShortStack/issues/70#issuecomment-360179306, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ AiGgjcAVV2cLktcdnZuyIE8CTm4w5cm2ks5tN1IHgaJpZM4RqzvD.

=ee-email-cont> This mail scanned by ARO IT mail secure system powered by PineApp Cybonet --ef.-1.dfa535fd3fb3d1ea08030b8ac3dcb74e.ef--

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/70#issuecomment-360188136, or mute the thread https://github.com/notifications/unsubscribe-auth/AGiXicj2x3MEqteML_6t65JjPENXxHb-ks5tN1gtgaJpZM4RqzvD .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

arazit commented 6 years ago

Will do. Thanks Tzahi

From: Mike Axtell [mailto:notifications@github.com] Sent: יום ד 24 ינואר 2018 19:56 To: MikeAxtell/ShortStack ShortStack@noreply.github.com Cc: Tzahi Arazi tarazi@volcani.agri.gov.il; Author author@noreply.github.com Subject: Re: [MikeAxtell/ShortStack] Cluster analysis using ShortStack (#70)

Definitely. You may just need to adjust the mincov if you feel the clustering is too aggressive. Sometimes each dataset takes a little tuning.

On Wed, Jan 24, 2018 at 11:19 AM, arazit notifications@github.com<mailto:notifications@github.com> wrote:

Hi Mike Thanks a lot. Can I use deep coverage and still for example detect Mirnases? Tzahi

-------- Original message -------- From: Mike Axtell notifications@github.com<mailto:notifications@github.com> Date: 1/24/18 18:00 (GMT+02:00) To: MikeAxtell/ShortStack ShortStack@noreply.github.com<mailto:ShortStack@noreply.github.com> Cc: Tzahi Arazi tarazi@volcani.agri.gov.il<mailto:tarazi@volcani.agri.gov.il>, Author < author@noreply.github.commailto:author@noreply.github.com> Subject: Re: [MikeAxtell/ShortStack] Cluster analysis using ShortStack (#70)

Hi Tzahi,

Yes, this is expected. Bascially, with deeper coverage but the same mincov, clusters grow larger and adjacent clusters merge together ... if there are two hotspots, they will be found as two separate clusters with low coverage data. But if you sequence deeper, it often is the case that there are a few reads found in between hotspots, which cause the two initial clusters to be merged into one.

Best, Mike

On Wed, Jan 24, 2018 at 2:06 AM, arazit notifications@github.com<mailto:notifications@github.com> wrote:

Dear Dr. Axtell, I have used ShortStack 3.8.3 to predict small RNA clusters. I noticed that the more small RNAs I analyze I get less clusters. This happened also when I use constant mincov and not relative mincov (rpm). I use foldsize = 1000. Do you have any idea why? Thanks, Tzahi

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/70, or mute the thread https://github.com/notifications/unsubscribe-auth/ <https://github.com/notifications/unsubscribe-auth/%0b> AGiXiR6ZqPjtLI4gLq9VC59LuRtXFm0Jks5tNtZjgaJpZM4RqzvD> .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/ <https://github.com/%0b> MikeAxtell/ShortStack/issues/70#issuecomment-360179306>, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ <https://github.com/notifications/unsubscribe-auth/%0b> AiGgjcAVV2cLktcdnZuyIE8CTm4w5cm2ks5tN1IHgaJpZM4RqzvD>.

=ee-email-cont> This mail scanned by ARO IT mail secure system powered by PineApp Cybonet --ef.-1.dfa535fd3fb3d1ea08030b8ac3dcb74e.ef--

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/MikeAxtell/ShortStack/issues/70#issuecomment-360188136, or mute the thread https://github.com/notifications/unsubscribe-auth/AGiXicj2x3MEqteML_6t65JjPENXxHb-ks5tN1gtgaJpZM4RqzvD .

-- Michael J. Axtell, Ph.D. Professor of Biology Penn State University http://sites.psu.edu/axtell

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/MikeAxtell/ShortStack/issues/70#issuecomment-360218946, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AiGgjRh-3ULgMZjC5YMZhWLAVxoa1Broks5tN26agaJpZM4RqzvD.