Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!nih-csl!lhc!usenet
From: usenet@nlm.nih.gov (usenet news poster)
Newsgroups: comp.ai.neural-nets
Subject: Re: possible NN application
Message-ID: <1990Sep27.024801.9755@nlm.nih.gov>
Date: 27 Sep 90 02:48:01 GMT
References: <26964@boulder.Colorado.EDU>
Reply-To: states@tech.NLM.NIH.GOV (David States)
Organization: National Library of Medicine, Bethesda, Md.
Lines: 45

In article <26964@boulder.Colorado.EDU> eesnyder@boulder.Colorado.EDU (Eric E. Snyder) writes:
>I am currently trying to develop a NN model to determine 
>a quantitative relationship between nucleotide sequences and 
>functional activities.  
>
>The data takes the following form:
>
>sequence		activity
>
>ATTTCT			2.3
>ACCTCC			3.0
>AAATCG 			2.5
>
>etc....
>
>I would like to develop a general model which can predict activity
>given a new sequence.  

What specifically did you have in mind?  The use of neural networks in
molecular sequence has been explored by a number of workers.  A couple
recent examples include:

Holley and Karplus, PNAS 86:152-6 (1989)
title: Protein secondary structure prediction with a neural network.

Qian N; Sejnowski TJ, J Mol Biol 202: 865-84 (1988)
title: Predicting the secondary structure of globular proteins using 
	neural network models.

Lukashin AV; Anshelevich VV; Amirikyan BR; Gragerov AI; Frank-Kamenetskii MD
J Biomol Struct Dyn 6: 1123-33 (1989)
title: Neural network models for promoter recognition.

Bohr H; Bohr J; Brunak S; Cotterill RM; Lautrup B; Norskov L; Olsen OH;
Petersen SB, FEBS Lett 241: 223-8 (1988)
title: Protein secondary structure and homology by neural networks. The
    alpha-helices in rhodopsin.

Kneller DG; Cohen FE; Langridge R, J Mol Biol 214: 171-82 (1990)
title: Improvements in protein secondary structure prediction by an enhanced
    neural network.

>Eric E. Snyder                            

David States