Copy Link
Add to Bookmark
Report
Neuron Digest Volume 09 Number 20
Neuron Digest Saturday, 25 Apr 1992 Volume 9 : Issue 20
Today's Topics:
neural nets and protein structure prediction
Re: neural nets and protein structure prediction
Re: neural nets and protein structure prediction
Re: neural nets and protein structure prediction
Re: HELP TO START
Software Patents
Re: soft patents
neural networks and neuroscience
Travel assistance to IJCNN?
Bridle e-mail address request
Send submissions, questions, address maintenance, and requests for old
issues to "neuron-request@cattell.psych.upenn.edu". The ftp archives are
available from cattell.psych.upenn.edu (128.91.2.173). Back issues
requested by mail will eventually be sent, but may take a while.
----------------------------------------------------------------------
Subject: neural nets and protein structure prediction
From: robert.h.gross@dartmouth.edu (Robert H. Gross)
Organization: Dartmouth College, Hanover, NH
Date: 14 Apr 92 18:45:10 +0000
Hi folks,
I am interested in identifying some recent papers that address the use
of neural nets in determining protein structure. Can someone please
post some references?
Thanks.
Bob Gross
Dartmouth College
bob.gross@dartmouth.edu
------------------------------
Subject: Re: neural nets and protein structure prediction
From: richard@frej.scs.uiuc.edu (Richard A. Goldstein)
Organization: School of Chemical Sciences Univ. of Illinois, Urbana IL
Date: 15 Apr 92 05:09:26 +0000
In article <1992Apr14.184510.9403@dartvax.dartmouth.edu> robert.h.gross@dartmouth.edu (Robert H. Gross) writes:
>Hi folks,
>
>I am interested in identifying some recent papers that address the use
>of neural nets in determining protein structure. Can someone please
>post some references?
>
>Thanks.
>
>Bob Gross
>Dartmouth College
>bob.gross@dartmouth.edu
We have been at work developing ways to use Hopfield Neural Networks to
predict protein tertiary structure. See:
Author = Mark S. Friedrichs and Peter G. Wolynes,
Title = Toward Protein Tertiary Structure Recognition by
Means of Associative Memory Hamiltonians,
Journal = Science,
Year = 1989,
Volume = 246,
Pages = 371-373
Author = Mark S. Friedrichs and Peter G. Wolynes,
Title = Molecular Dynamics of Associative Memory Hamiltonians
for Protein Tertiary Structure Recognition,
Journal = Tetra. Comp. Meth.,
Year = 1991,
Volume = 3,
Pages = 175-190
Author = Mark S. Friedrichs and Richard A. Goldstein
and Peter G. Wolynes,
Title = Molecular Dynamics of Associative Memory Hamiltonians
for Protein Tertiary Structure Recognition,
Journal = J. Mol. Biol.,
Year = 1991,
Volume = 222,
Pages = 1013-1034
Author = Richard A. Goldstein and Zan Luthey-Schulten and
and Peter G. Wolynes,
Title = Optimal Protein Folding Codes from Spin-Glass Theory,
Journal = PNAS,
Year = 1992,
Note = in press (Next month, I think)
Also see the references contained in these papers, especially the JMB paper.
Other work using feed-forward neural networks for tertiary structure
prediction has been done by:
Author = H. Bohr and J. Bohr and S. Brunak and R. M. J. Cotterill
and B. Lautrup and L. Norskov and O Olsen
and S. Petersen,
Year = 1990,
Title = A Novel Approach to Prediction of the 3-dimensional
structure of Protein Backbones by Neural Networks,
Journal = FEBS Lett.,
Volume = 261,
Pages = 43-46
Author = G. L. Wilcox and M. Poliac and A. Brugge and
M. N. Liebman,
Journal = Tetra. Comp. Meth.,
Title = Neural Network Analysis of Protein Tertiary Structure,
Volume = 3,
Pages = 191-211
Probably more work has been done on secondary-structure prediction:
Author = N. Qian and T. J. Sejnowski,
Year = 1988,
Journal = J. Mol. Biol.,
Title = Predicting the secondary sequence of globular proteins
using neural network models,
Volume = 202,
Pages = 865-884
Author = L. H. Holley and M. Karplus,
Year = 1989,
Title = Protein Secondary Structure Prediction with a
Neural Network,
Journal = PNAS,
Volume = 86,
Pages = 152-156
Author = H. Bohr and J. Bohr and S. Brunak and R. M. J. Cotterill
and B. Lautrup and L. Norskov and O Olsen
and S. Petersen,
Year = 1988,
Title = Protein Secondary Structure and Homology by Neural
Network--The $\alpha$-helices in Rhodopsin,
Journal = FEBS Lett.,
Volume = 241,
Pages = 223-228
Author = D. G. Kneller and F. E. Cohen and R. Langridge,
Title = Improvements in Protein Secondary Structure Prediction
By an Enhanced Neural Network,
Journal = J. Mol. Biol.,
Volume = 214,
Year = 1991,
Pages = 171-182
Hope this helps.
Richard Goldstein
richard@frej.scs.uiuc.edu
------------------------------
Subject: Re: neural nets and protein structure prediction
From: lambert@hydrogen.oscs.montana.edu (Chris Lambert)
Organization: Montana State University
Date: 15 Apr 92 13:04:55 +0000
There are more recent articles but this reference was hanging
around:
Qian, N. & Sejnowski, T. J., Predicting the Secondary Structure of
globular Proteins Using Neural Network Models. (1988) J. Mol. Biol.
202, 865-884.
Chris Lambert
Montana State University
108 Gaines Hall
Bozeman, MT. 59717
(406) 994-1964
lambert@hydrogen.oscs.montana.edu
------------------------------
Subject: Re: neural nets and protein structure prediction
From: steeg@cs.toronto.edu ("Evan W. Steeg")
Organization: Department of Computer Science, University of Toronto
Date: 15 Apr 92 16:33:21 +0000
Here are the ones I have in my bibtex bibliography file right now:
This first one is the *first* paper on the topic. Qian and Sejnowski
used a backprop net and NetTalk-like "local window" approach to
predict secondary structure.
@article{qian-sejnowski-88,
title="Predicting the secondary structure of globular proteins
using neural network models",
author="Qian, N. and Sejnowski, T. J.",
year="1988",
journal="Journal of Molecular Biology",
volume="202",
pages="865-884",
annote="
The best existing method for predicting the secondary structure
of a globular protein is a neural network:
Results also indicate that only marginal improvements
on our performance will be possible with local methods.
Tertiary (3-D) structure is a much more difficult problem
for which there are no good methods.
- -- Terry"
}
The next few are also *secondary* structure prediction papers, and
offer some slight changes and extensions to the Qian/Sejnowski work.
@article (holley-karplus-89,
Key = "Holley",
author = "Holley, L. H. and Karplus, M.",
title = "Protein Secondary Structure Prediction with a Neural Network",
year = "1989",
journal = "Proceedings of the National Academy of Science USA",
volume = "86",
pages = "152-156"
)
@techreport ( fogelman-90,
key = "Fogelman-Soulie" ,
author = "Mejia, C. and Fogelman-Soulie, F." ,
title = "Incorporating Knowledge in Multilayer Networks: The Example
of Proteins Secondary Structure Prediction",
type = "Rapport de Recherche" ,
institution= "Laboratoire de Recherche en Informatique, Univ. de Paris-Sud",
year = "1990"
)
@article (bohr-et-al-88,
Key = "Bohr",
author = "Bohr, H. and Bohr, J. and Brunak, S. and
Cotterill, R. M. and Lautrup, B. and Norskov, L. and
Olsen, O. H. and Petersen, S. B.",
title = "Protein Secondary Structure and Homology by Neural
Networks: The Alpha-Helices in Rhodopsin",
year = "1988",
journal = "FEBS Letters",
volume = "241",
pages = "223-228"
)
@article{Kneller-at-al-90,
author = "D. G. Kneller
and F. E. Cohen
and R. Langridge",
title = "Improvements in protein secondary structure prediction by an
enhanced neural network",
journal = "J. Mol. Biol.",
volume = "214",
pages = "171-182",
year = "1990"}
The group listed below came up with a very interesting result: A simple
Bayesian one-shot classifier, embodying the *very* unrealistic assumption
of statistical independence between adjacent and nearby amino acid
residues, worked nearly as well as the most sophisticated methods. This,
I think, tells us quite forcefully that the 60-70% predictive accuracy
rate "wall" will never be breached without discarding the local window
approach and trying to incorporate more global, and tertiary, structural
information into secondary structure prediction.
@techreport ( lapedes-prot,
key = "lapedes",
author = "Stolorz, P. and Lapedes, A.~S. and Xia, Y",
title = "Predicting Protein Secondary Structure Using Neural
Net and Statistical Methods",
type = "Technical Report",
number = "LA-UR-91-15",
institution= "Los Alamos National Laboratory",
year = "1991",
bibdate = "Tue Mar 1 19:21:22 1988" ,
keywords= "nonlinear signal processing, backpropagation"
)
The Shavlik group at Wisc. has been working on some interesting ways to
move between neural network and rule-based approaches to adaptive pattern
recognition. They have applied their methods to several tasks in
computational molecular biology, including promoter recognition and
protein secondary structure prediction. Unfortunately, I seem to have
lost 2 of the references, but here's one:
@unpublished (towell-shavlik-kbann,
AUTHOR = "Towell, G.~G. and Shavlik, J.~W.",
YEAR = "1991",
TITLE = "The Extraction of Refined Rules from Knowledge-Based
Neural Networks",
NOTE = "Submitted to {\em Machine Learning}"
)
The next several are about *tertiary* structure prediction. The Wolynes
paper below is not about neural nets per se, but describes a NN-like
method for classifying/recognizing/predicting protein structures. (The
Wolynes group has a new, 1992, paper out as well, but I don't have the
reference handy).
@article ( wolynes-89,
key = "Wolynes" ,
title = "Toward Protein Tertiary Structure Recognition by
Associative Memory Hamiltonians",
author = "Friedrichs, M. S. and Wolynes, P. G." ,
journal = "Science" ,
year = "1989" ,
volume = "246" ,
pages = "371-373"
)
@article (bohr-et-al-90,
Key = "Lautrup",
author = "Bohr, H. and Bohr, J. and Brunak, S. and
Cotterill, R. M. J. and Fredhom, H. and Lautrup, B. and
Petersen, S. B.",
title = "A Novel Approach to Prediction of the 3-Dimensional
Structures of Protein Backbones by Neural Networks",
year = "1990",
journal = "FEBS Letters",
volume = "261",
pages = "43-46"
)
Our work, described in the papers referenced below, combines partial
tertiary structure prediction with classification, and is an attempt at
combining bottom-up (sequence -> 2-ary struc -> 3-ary struc) information
with top-down (3-ary struc -> secondary struc) information in a unified
global constraint satisfaction approach.
@INPROCEEDINGS{greller-steeg-salemme-91a,
AUTHOR = {Greller, L. D. and Steeg, E. W. and Salemme, F. R.},
TITLE = {Neural Networks for the Detection and Prediction of 3D
Structural Domains in Proteins},
BOOKTITLE = {Proceeedings Eighth International Conference on
Mathematical and Computer Modelling},
YEAR = {1991},
ADDRESS = {College Park, Maryland},
MONTH = {April},
ANNOTE = {A newer version has been submitted to {\em Science}.}
}
I also review some of these approaches and the computational issues in
the following (which focuses on RNA structure, however). There is at
least one other chapter in the book that describes protein structure
prediction with NNs, but, again, I don't have the names. Sorry.
@incollection ( steeg-bookchapter,
author = "Steeg, E.~W.",
year = "1992",
title = "Neural Networks, Adaptive Optimization, and
{RNA} Secondary Structure Prediction",
booktitle = "Artificial Intelligence and Molecular Biology",
editors = "Hunter, L.",
publisher = "AAAI Press",
note = "In Press"
)
That's all for now.
-- Evan
Evan W. Steeg (416) 978-5182 steeg@ai.toronto.edu (CSnet,UUCP,Bitnet)
Dept of Computer Science steeg@ai.utoronto (other Bitnet)
University of Toronto, steeg@ai.toronto.cdn (EAN X.400)
Toronto, Canada M5S 1A4 {seismo,watmath}!ai.toronto.edu!steeg
------------------------------
Subject: Re: HELP TO START
From: M_FATTAH%EGFRCUVX.BITNET@mitvma.mit.edu
Date: Sat, 18 Apr 92 14:02:00 +0000
[[ Editor's Note: Another, not infrequent, request for "beginner's"
information. Since this person seems to have a particular field in mind,
I hope some readers will be able to help. Don't forget to email to
Neuron Digest as well as to M. Fattah, so all may benefit. Alas, I have
not kept up with the current crop of introductory texts or articles, and
still tend to recommend the (now slightly dated) PDP series by Rumelhart
and McClelland. I, for one, would appreciate any other general text
recommendations (especially for college undergraduate level) or
reassurance the the R&M books are still useful. -PM ]]
My field of interset is the vision science and its engineering
applications in the illuminating engineering. I would like to obtain the
names for Tutorial books and papers on that Neural Networks subject.
Mohamed A. Fattah
------------------------------
Subject: Software Patents
From: Ken Laws <LAWS@ai.sri.com>
Date: Sat, 18 Apr 92 22:48:01 -0800
> It now seems that as of 21 January, 1992, IBM has been granted a patent
> on the idea of language binding. Although they are willing to licence the
> patent, the situation would seem to be that IBM has it within its power
> to invalidate a substantial number of standards.
I have not seen the patent, and don't know precisely what its claims are.
If it truely claims all applications of language bindings, the patent
should be trivial to overturn. Just cite any or all of the graphic
standards that use such bindings. Since they are fully documented and
published, they unquestionably establish prior art and obviousness. (I'm
assuming that the IBM patent was filed fairly recently. If it dates back
many years, it may have indeed have the prior claim.)
This really isn't the forum in which to debate patent law. Usenet has
several forums for that. I am responding here only because the issue was
raised here, and because the posting has a "Chicken Little" quality that
disturbs me. Patents are granted for applications, not ideas, and
usually for very specific uses of technology. I'm very suspicious of
people who want me to petition the king about our current danger, but who
fail to present the evidence. What are the specific claims in the IBM
patent, and why do you believe they could restrict prior art?
If members of this list wish to continue the discussion, I hope you will
keep in mind that neither copyright nor trade secret protection are of
any use for neural networks. As Tom Schwartz has pointed out to me, it
is easy to use one neural chip to train another, arriving at an entirely
new set of coefficients that perform the same function.
-- Ken Laws
------------------------------
Subject: Re: soft patents
From: me@suzuka.u-strasbg.fr (Michel Eytan LILoL)
Date: Sun, 19 Apr 92 15:19:42 +0000
Fellow netters,
I believe the situation (illustrated by one more example in Marcus Speh's
message) is properly scandalous. I include the call to join the League for
Programming Freedom (hope there is no patent on this!...)
[[ Editor's Note: I have not included the submitted call, since it is
quite long and really does fall outside the purview of this Digest.
However, the League for Programming Freedom is quite willing to send you
information about their organization, recent "look and feel" lawsuits,
software patents, etc. They are an advocacy group, not strictly a legal
clearing house, however. Richard Stallman, president of the League, is
best known as the proginator of GnuEmacs and the rest of the Gnu software
group. If you have any questions, please phone the League at (617)
243-4091 or send Internet mail to league@prep.ai.mit.edu. ]]
------------------------------
Subject: neural networks and neuroscience
From: MCCAINKW@DUVM.BITNET (Kate McCain)
Date: 21 Apr 92 11:11:25 +0000
I am finishing the first of a series of papers on the interdisciplinary
nature of neural networks research -- focusing on information needs &
information use and the potential problems libraries face in supporting
NN research activities. One strong finding, so far, is that the degree of
participation of neuroscientists and other biologists (and psychologists)
seems to be much more visible in USE patterns (citations in articles in
journals like _Neural Networks_ and _Neural Computation_) than in
publication patterns (journals publishing many articles with key words in
titles or articles citing "classic" NN-related work). In publication
choice, and in a survey that Michael Rappa and Karl DeBackere did in
1988, about 3/4 of the journals and respondents, respectively,
represented the physical sciences and engineering.
One obvious interpretation is that the recent growth in NN research has
been primarily in PS&E -- that there are so many more people writing, and
writing in that journal literature that bio/psych work is simply a
smaller proportion.
Other, partial, interpretations are that there is a "one-way flow" of
information from bio/psych to PS&E; that is (to quote a friend of mine)
"they (PS&E NN researchers) use our data but haven't told us anything we
can use yet" or that bio/psych authors are as likely to publish in PS&E
journals as in their own neuroscience or psychology journals (or the
"cross-over"
journals like _Biological Cybernetics_) perhaps because that way they
reach the audience of interest.
I am very interested in comments from any netters who read or participate
in neural networks research (defined here in the sense of "artificial
neural systems" or "neurobiologically inspired computer architectures."
Is this research relevant to neuroscience? Do you read/publish in the
PS&E journals or the new NN-related journals? Is it important that your
library maintain subscriptions to PS&E journals that publish NN-related
research?
All insights gratefully accepted.
Kate McCain MCCAINKW@DUVM
Associate Professor mccainkw@duvm.ocs.drexel.edu
College of Information Studies
Drexel University "Die Gedankenexperimente sind Frei!!"
Philadelphia PA 19104
------------------------------
From: cc!eesmbis@gateway.iitb.ernet.in
Date: Fri, 24 Apr 92 19:45:00 +0700
Subject: Travel assistance to IJCNN?
[[ Editor's Note: Time is quite short for this fellow and the distance is
quite large. I don't know of organized travel grants for paper
presentations, since conferences assume the individual's organization
will pick up the tab. This request is similar to one from a former East
Bloc colleague a few months ago, whose government (after swicthing from
Communism) lacked the funds for scientific travel. Perhaps someone can
offer some insight or direction? -PM ]]
Dear sir,
I am a research scholar in the Dept. of Electrical Engg. at Indian
Institute of Technology,Bombay, INDIA. My paper is accepted for
presentation at IJCNN-92 (Baltimore) and I am looking for some
organization who can provide me financial support for attending the
conference (mainly financial support towards travel). Could you please
give any information regarding any organization which can provide such
financial support. Thanking you.
Sanjay Bhandari.EE Dept.,IIT Powai, Bombay.400 076.INDIA.
------------------------------
Subject: Bridle e-mail address request
From: juanma@hal.ugr.es
Date: Fri, 24 Apr 92 16:53:05 -0300
[[ Editor's Note: On principle and with respect to the privacy of Neuron
Digest subscribers, I do not give out addresses of subscribers or even
whether a particular person is a subscriber. If someone else can offer
assistance, however, please do... -PM ]]
Hello,
First, let me introduce myself, my name is J. Manuel Lopez Soler from
the Electronic Department of the University of Granada (Spain).
I'm trying to reach Dr. J.S. Bridle by e-mail, but I don't have his
address. He works at Speech Research Unit, Royal Signals and Radar
Establishment, Malvern, Great Britain.
Could you be (if you have it) so kind as to send me Dr. Bridle e-mail address?
Thanks in advance, best regards
Juan Manuel.
------------------------------
End of Neuron Digest [Volume 9 Issue 20]
****************************************
