is it posiible to run topcons docker only for topcons

[naveen584]

is it posiible to run topcons docker only for topcons:

i just want to run topcons docker only for topcons to save some timem i dont want to run this docker for other like SPOCTOPUS,OCTOPUS etc so is there is any way can you please tell, it will be very helfful for me.

[nanjiangshu]

Hi @naveen584, TOPCONS is a consensus predictor which means it will first make predictions for individual predictors such as SCAMPI, SPOCTOPUS, Phobius etc and then combine these predicted topologies into one by a Hidden Markov Model. So you can not get TOPCONS prediction without running individual predictors. Generally speaking, the most time consuming part of the whole pipeline is sequence profile making, which means you won't save much time to run just one of the MSA based predictions.

[naveen584]

ok thanks one more thing if you don't mind may i know the configuration of your machine where your topcons are running because it runs fast compared to my machine , for me it takes 3 hours to run for 30 sequences. <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail> Virus-free. www.avast.com <https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

…

On Wed, Sep 16, 2020 at 6:20 PM Nanjiang Shu ***@***.***> wrote: Hi @naveen584 <https://github.com/naveen584>, TOPCONS is a consensus predictor which means it will first make predictions for individual predictors such as SCAMPI, SPOCTOPUS, Phobius etc and then combine these predicted topologies into one by a Hidden Markov Model. So you can not get TOPCONS prediction without running individual predictors. Generally speaking, the most time consuming part of the whole pipeline is sequence profile making, which means you won't save much time to run just one of the MSA based predictions. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#11 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKY4HSCVIGR2USEWVCMV2X3SGCYAHANCNFSM4ROYY27A> .

[nanjiangshu]

Hi @naveen584, the TOPCONS web-server has a distributed backend with multiple nodes, at moment there are in total more than 120 CPU cores. There are also thousands of jobs submitted to the web-server everyday so your job might be finished slower or faster depending on the load of the server. In addition, if your sequence has a homologous family in Pfam, it takes much shorter time to build the sequence profile and it usually takes less than 30 seconds to predict one sequence. Otherwise, a full blast searching will be run and it may take more than 10 minutes or even hours to predict one sequence. So the speed various a lot depending one your query.

[arneelof]

It all depends on if you have a hit in the "preprocesss pfam database" or not (see paper). Without a hit it takes 5-10 minutes per sequence. i.e. more or less the same as you. We also have caching - so if someone has run your sequences before it is fast.. Yours Arne

…

----------------------------------------- Arne Elofsson Science for Life Laboratory Tel:+46-(0)70 695 1045 Stockholm University http://bioinfo.se/ Box 1031, Email: arne@bioinfo.se 17121 Solna, Sweden Twitter: https://twitter.com/arneelof Scholar: http://scholar.google.se/citations?user=s3OCM3AAAAAJ ORCID: 0000-0002-7115-9751 On Wed, Sep 16, 2020 at 3:49 PM Naveen Kumar <notifications@github.com> wrote:

ok thanks one more thing if you don't mind may i know the configuration of your machine where your topcons are running because it runs fast compared to my machine , for me it takes 3 hours to run for 30 sequences. < https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail > Virus-free. www.avast.com < https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail > <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2> On Wed, Sep 16, 2020 at 6:20 PM Nanjiang Shu ***@***.***> wrote: > Hi @naveen584 <https://github.com/naveen584>, TOPCONS is a consensus > predictor which means it will first make predictions for individual > predictors such as SCAMPI, SPOCTOPUS, Phobius etc and then combine these > predicted topologies into one by a Hidden Markov Model. So you can not get > TOPCONS prediction without running individual predictors. Generally > speaking, the most time consuming part of the whole pipeline is sequence > profile making, which means you won't save much time to run just one of the > MSA based predictions. > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub > < #11 (comment)>, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AKY4HSCVIGR2USEWVCMV2X3SGCYAHANCNFSM4ROYY27A > > . > — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#11 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AA3LGSJSCMJ2GBNTDVW2KRLSGC64NANCNFSM4ROYY27A> .