Please consider submission to the AAAI symposium described below. Note that
abstracts are due by e-mail on October 30.
Thank you.
Marco
Marco Valtorta, Associate Professor and Undergraduate Director
Department of Computer Science internet: mgv@usceast.cs.sc.edu
University of South Carolina tel.: (1)(803)777-4641 fax: -3767
Columbia, SC 29208, U.S.A. http://www.cs.sc.edu/~mgv/ tlx: 805038 USC
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Symposium:
PREDICTIVE TOXICOLOGY OF CHEMICALS: EXPERIENCES AND IMPACT OF AI TOOLS
Stanford University (CA), March 22-24, 1999
within the American Association for Artificial Intelligence Spring
Symposium Series
AI and related techniques play a major role in toxicity prediction. The
goal of computational toxicity prediction is to describe the relationship
between chemical properties, on the one hand, and biological and
toxicological processes, on the other. This symposium will highlight the
potential of different AI approaches, either individually and combined, for
computational toxicity prediction.
Success in this research depends on the contribution of experts from
different areas, and we invite participation from researchers in all
related fields. We welcome AI researchers who have applied learning
techniques to domains outside toxicity prediction and are in search of new
areas.
Some of the questions to be addressed in the symposium are:
- How do we represent chemical information? Several methods have been
proposed. Are they equivalent? How do we evaluate them? Are results from
different experiments reproducible?
- How can machine learning (including ANN, fuzzy logic, GA, ILP, ...)
techniques be used? AI tools have yet to be fully evaluated in this
domain. Which techniques are better for toxicity prediction, especially
given our changing understanding of toxicology? Are hybrid approaches
better?
- Are current experimental data sets sufficient for AI techniques? Do they
have sufficient accuracy? How do we take advantage of existing data sets?
Can we use techniques from data mining and reasoning under uncertainty?
To achieve a common background among both computer scientists and chemists,
there will be short introductory presentations on the state of the art in
computational techniques, machine learning, chemical descriptors, and
toxicological prediction. The rest of the sessions will include
presentations (oral and poster) with a discussion on the open problems.
Submission information
Potential participants should submit an abstract describing work in
progress, completed work, positions, or even open questions for discussion.
Abstracts should be submitted electronically to Giuseppina Gini
(gini@elet.polimi.it), including title, author's name(s), affiliation,
mailing address, e-mail, phone and fax numbers.
Deadline for abstract submission is October 30, 1998. Notification of
acceptance will be given by November 14. Participants may be invited to
submit a longer version of their paper. All contributions will be collected
in working notes. Some financial assistance is available for student
participation.
Further information and format for submissions will be posted on a WWW home
page at: http://www.elet.polimi.it/AAAI-PT.
See also the page of the American Association of Artificial Intelligence:
http://www.aaai.org/
Organizing committee
Giuseppina C. Gini, (chair)
Dipartimento di Elettronica e Informazione, Politecnico di Milano,
piazza L. da Vinci 32, 20133 Italy
Telephone: (+39) 02-23993626; FAX: (+39) 02 - 23993411;
Email address: gini@elet.polimi.it
WWW Homepage: http://www.elet.polimi.it/people/gini/index.html
Alan R. Katritzky, (cochair), University of Florida, Gainesville, FL
(katritzky@chem.ufl.edu)
Emilio Benfenati, Istituto Mario Negri, Milan, Italy (benfenati@irfmn.mnegri
.it)
Daniel L. Boley, University of Minnesota, Minneapolis, MN (boley@cs.umn.edu)
Adolf Grauel, University of Paderborn, Soest, Germany
(grauel@ibm5.uni-paderborn.de)
Marco Valtorta, University of South Carolina, Columbia, SC
(mgv@usceast.cs.sc.edu)
Yin-tak Woo, EPA, Washington, DC
(WOO.YINTAK@epamail.epa.gov)
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
GENERAL INFORMATION ABOUT THE
AAAI 1999 Spring Symposium Series
to be held in Stanford University, Stanford, Ca
The general SCHEDULE for the symposium series will be:
22 March 1999
9:00 Session
10:30 Coffee Break
11:00 Session
12:30 Lunch
2:00 Session
3:30 Coffee Break
4:00 Session
5:30 End of Sessions
evening Opening Reception
23 March 1999
9:00 Session
10:30 Coffee Break
11:00 Session
12:30 Lunch
2:00 Session
3:30 Coffee Break
4:00 Session
5:30 End of Sessions
evening Plenary Session
24 March, 1999
9:00 Session
10:30 Coffee Break
11:00 Session
12:30 End of Symposium Series
PLENARY SESSION:
In the evening plenary session, open to the public, representatives from
each symposium talk about their symposium. and present some of the issues
behind.
SUMMARY OF DEADLINES:
October 30 Abstracts are due to symposia chairs
November 13 Acceptance/rejection notices are mailed out by symposia chairs
December 18 AAAI will mail the registration brochure
Jan 15, 1999 participants send camera-ready copy of theirpapers, A/V
requests, and "Permissions to Distribute" forms to the chairs.
February 5 Invited participants registration deadline
February 26 Final (open) registration deadline
March 22-24 Spring Symposium Series, Stanford University
The logistics, including room set-up, audio visual, signs, printing of
working notes, catering, registration, plenary session, etc. will be
organized by AAAI.
WORKING NOTES AND TECHNICAL REPORTS:
The Working Notes given to participants are for discussion purposes
only. Participants can resubmit elsewhere a final version of their
presentations.
FORMAT OF THE FINAL VERSION
The working notes should be double column, with at least 3/4" of white
space on all sides of 8 1/2" by 11" paper. (If anyone uses A4 paper, they
should not center on the A4 paper, but instead center on an 8 1/2" by 11"
area measured from the top left corner of the A4 paper.) You can use
the style guides of any of the recent AAAI or IJCAI conferences, as long
as they leave the correct amount of white space.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
LAST NEWS FROM OUR SYMPOSIUM ON PREDICTIVE TOXICOLOGY
The symposium will be articulated into ten sections, each devoted to an
open problem, eventually introduced by a tutorial-style presentation, or
eventually organized as a round-table. There is no problem with presenting
work-in-progress in the symposia, rather than just finished work.
We want to investigate how AI and related techniques can play a major role
in toxicity prediction. The goal of computational toxicity prediction is to
describe the relationship between chemical properties, on the one hand, and
biological and toxicological processes, on the other. This symposium will
highlight the potential of different AI approaches, either individually and
combined, for computational toxicity prediction.
Some of the finalized sessions will deal with the following questions:
What do we mean by toxicity and toxicity prediction?
There will be a general introductory presentation on the current issues in
the prediction of toxicity by A. Richard, EPA, USA.
How do we represent chemical information? Several methods have been
proposed. Are they equivalent? How do we evaluate them? Are results from
different experiments reproducible?
A tutorial speach will be on the use of molecular descriptors for the
prediction of properties of chemicals, by A. Katritzky, University of
Florida, USA.
How machine learning (including ANN, fuzzy logic, GA, ILP, ...)
techniques have been used so far? AI tools have yet to be fully evaluated
in this domain. Which techniques are better for toxicity prediction,
especially given our changing understanding of toxicology? Are hybrid
approaches better?
Are current experimental data sets sufficient for AI techniques? Do they
have sufficient accuracy? How do we take advantage of existing data sets?
Can we use techniques from data mining and reasoning under uncertainty?
How the classical correlation studies done in QSAR can be related to the
AI studies? What are the weak points of different approaches?
A tutorial speach will be on The relations between AI representations and
probability.
How to use the prediction of toxicity (risk assessment, priorities
definition, mixture toxicity, ...)?
The symposium accepts people, not written submissions or presentations. If
you want to participate without giving presentations you may ask so, only
submitting a short curriculum. More than specific talks on the work done,
the symposium will attempt to find the work to be done, the problems and
the possible solutions.
At present, some of the possible themes emerging from submitted
contributions are (in mixed order)
A Comprehensive Approach to Argumentation.
The Use of Multimodel QSAR Approaches
Toxicity Predictions and Quantitative Shape-Activity Relations (QShAR)
Based on Shape Analysis of Electron Density Charge Clouds
Rule generation by means of lattice theory
Results and experiences in some EU projects in evaluating toxicity
Prediction of carcinogenity using hybrid systems
On the prediction of toxicity of mixtures
A hybrid approach to risk assessment for multiple pathway exposures
The needs in toxicity evaluation for a phytochemical company.
Prediction of pesticides toxicity
Inductive Logic Programming systems for toxicity prediction
.....
PUBLICATION
We would like to turn our Working Notes into a AAAI Technical Report that
will continue to be available from AAAI. This option requires that each
participant sign a Permission to Distribute Form BEFORE the working
notes are put together. Authors still will retain all rights to their
submissions. We will provide you this form.
AAAI Support for Students
EACH SYMPOSIUM WILL RECEIVE SOME MONEY TO SUPPORT GRADUATE STUDENT TRAVEL
We will decide how many students get the money as well as how much each
student gets. All the accepted students will submit their receipts for
reimbursement directly to AAAI.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Giuseppina Gini
address: Dip. di Elettronica e Informazione, Politecnico di Milano
piazza L. da Vinci 32, I-20133 MILANO
fax: (+39) 2-2399.3411
phone: (+39) 2-2399.3626e-mail - gini@elet.polimi.it
home page: http://www.elet.polimi.it/people/gini/
http://www.elet.polimi.it/AAAI-PT for AAAI Spring Symposium on Predictive
Toxicology