From shight at purdue.edu Mon Mar 8 16:43:08 2004 From: shight at purdue.edu (suzie.k.hight.1) Date: Mon Mar 8 21:43:51 2004 Subject: [Jpred-list] explanation of output Message-ID: Hello, I have a very simple question: Could anyone give me a brief and basic explanation of what the different JPred outputs mean? For instance, in the .pdf file, there is an alignment of the outputs from jnet, jhmm, jalign, jpssm, jfreq, and jpred. Are these different methods, or outputs from different stages from the jnet prediction? I apologize if this has been answered in the past - for some reason I am not able to access the Jpred mailing list archives (I get the message "You don't have permission to access /mailman/public/ on this server"). Thank you very much! Suzie Hight ________________________________ Suzie Hight Dept. of Biology Purdue University From jon at compbio.dundee.ac.uk Wed Mar 10 11:52:56 2004 From: jon at compbio.dundee.ac.uk (Jonathan Barber) Date: Wed Mar 10 11:53:02 2004 Subject: [Jpred-list] explanation of output In-Reply-To: References: Message-ID: <20040310115256.GA21282@flea.compbio.dundee.ac.uk> On Mon, Mar 08, 2004 at 04:43:08PM -0500, suzie.k.hight.1 wrote: > Hello, > I have a very simple question: Could anyone give me a brief > and basic explanation of what the different JPred outputs mean? For > instance, in the .pdf file, there is an alignment of the outputs from > jnet, jhmm, jalign, jpssm, jfreq, and jpred. Are these different methods, > or outputs from different stages from the jnet prediction? These are the the predictions from the different neural networks that are part of Jnet (the program Jpred uses to do the actual prediction). Below is a list of the outputs, and what data produces them. * jpssm - The PSI-BLAST profile * jhmm - The HMMER profile of the alignment * jalign - The multiple sequence alignment * jfreq - The multiple sequence alignment weighted by BLOSUM 62 scores * jnet - The results of running a NN on the previous outputs where they don't agree. * jpred - The final output based upon the output of jnet with filtering to remove unlikely secondary structure such as helicies that are too short. > I apologize if this has been answered in the past - for some reason I am No problem, this should probably be placed in the FAQ. > not able to access the Jpred mailing list archives (I get the message "You > don't have permission to access /mailman/public/ on this server"). > Thank you very much! This is a problem with the mailing list archieves, I've mentioned it our sys admin, who will hopefully fix it soon. > Suzie Hight -- Jon