From:	mn1 - at - helix.nih.gov (M. Nicklaus)
Date:	Fri, 12 May 95 15:39:37 -0400
Subject:	Summary: Personal Structure/Bioassay/Inventory DB's

Dear CCL'ers,

This is the summary of the responses I received to my question
on Personal Structure/Bioassay/Inventory Databases.  My question as
well as the answers have been slightly edited for the sake of
brevity.  I have also included short excerpts from a magazine
article I was made aware of, and which might be interesting in this
context.  It appears that there are a number of exciting new
developments going on in this field.  Thanks to all who responded.

Marc C. Nicklaus

======================================================================

QUESTION:
=========
(asked on May 4, 1995)

We'd like to ask everyone [about] ...personal and project-oriented
chemical databases [for] typically ...up to a few hundred or maybe 1000
[compounds].  Such a database (system) would ideally
- allow storage and search of 2D chemical structures, numerical data
  such as bioassay results and inventory data, textual data such as
  comments and planned steps in a project, reference data such as
  literature references etc.  A really *good* system would allow
  intelligent handling of semi-quantitative data such as "IC50 > 100uM"
  for the purposes of queries, sorting etc.;
- allow "upward linking", i.e. to molecular modeling (programs)....
- allow "downward linking", i.e. to word processing, DTP and similar
  programs for the purpose of manuscript and presentation preparation.
  It should at least allow printing of publication-quality output such
  as tables containing 2D structures etc., but should ideally allow to
  create, e.g., PostScript files and/or export output to the common
  word processor formats such as WordPerfect, Word etc.
- be available on, and allow seamless data transfer between, the
  platforms/OS's Unix, PC's (DOS(?) / Windows / OS/2), Mac's;
- be easy to install, maintain, and use even for a non-database expert;
  in particular, the database definition (number, order, name of columns,
  type of data etc.) should be easily changeable as the project evolves;
- be as low-cost as possible, or even an outright public domain program.

======================================================================

ANSWERS:
========

From: "Craig Shelley" 

Entropy from SoftShell will ship in late May.  It is intended for personal or
small project databases and fits most of your requirements.  The price is $599
business, $399 academic or government, $99 student.  It requires our structure
drawing programs, ChemIntosh or ChemWindow.
- allow storage and search of 2D chemical structures, numerical data....
>>It should satisfy these needs, or at least most of them - it is not entirely
clear to me based on this comment.
- allow "upward linking", i.e. to molecular modeling (programs)....
>> You can save structures as MDL molfiles that can be transferred to many
molecular modeling packages.  It also supports SCF, a nonproprietary chemistry
standard that we are glad to share with others.
- allow "horizontal linking", i.e. to other, larger-scale database....
>> It cannot do this in the first release.  MDL SDfiles can be imported into
Entropy.
- allow "downward linking", i.e. to word processing, DTP....
>> Yes.  This is what we specialize in.
- be available on, and allow seamless data transfer between, the platforms....
>> Available for Windows and Macintosh.  The Entropy file format is the same for
both platforms.  The files can be transferred or placed on servers accessible to
both Window and Macintosh computers.
- be easy to install, maintain, and use even for a non-database expert....
>> We have tried to make this product simple.  We had 400 Entropy beta testers
that were very helpful.  Entropy excels at creating databases without formatting
too.
- be as low-cost as possible, or even an outright public domain program.
>> It's not free.  No matter how simple a database program is, you invest time
in creating the data for it.  Would you really want an unsupported product that
you invest a significant amount of time in?

For information on Entropy or our spring catalog of chemistry products, send
email to info # - at - # softshell.com.  Please don't send general information
requests to
me.

Craig Shelley, President
SoftShell

----------------------------------------------------------------------

From: James Dill 

I'm very interested in your posting regarding a personal chemical database.
It comes at just the right time, as I am working at Cambridge Scientific
Computing (now known as CambridgeSoft) on a new project to develop such a
system for Windows.  Although it will probably be called ChemFinder, it will
have little or nothing in common with our Mac program by that name, in case
you have knowledge of that one.

I plan to study your list of requirements closely, and will have questions
which I hope you are available to answer by e-mail.  I'd like to keep abreast
of any suggestions you hear from respondents.  My main question at the moment
is:  which system now available comes closest to what you need, and where does
it fall short?

We're committed to delivering what the community wants.  You're invited to
help design it.  Thanks in advance for your input.

  -- Jim Dill
      CambridgeSoft
      jdd $#at#$ camsci.com

----------------------------------------------------------------------

From: D.Winkler - at - chem.csiro.au (Dr. Dave Winkler)

We have been doing this in several ways for a few years.  The best solution
we found is the Tripos Unity/Sybyl database/modelling system.  It is
available on multiple platforms, allows fast substructure searches which
can link easily to data stored in Oracle or Ingress, links transparently to
the excellent Sybyl modelling package.  There is a new product called
Unison which combines all these functions in a very user-friendly manner.
I can recommend this approach.  Alternatives such as MDL's Isis are too
expensive for us (>100K last we heard) although their IsisBase database for
PC/Mac may do most of what you want if you have only 1000 or so compounds,
don't require a completely integrated solution like Unity/Unison and don't
want to spend much.

Our system is an SGI Indigo IIxl linked via TCP/IP to a network of Macs.
We run Oracle, Sybyl and Unity on the Indigo and connect from the Macs
using MacX.  As Unison is further developed, we will switch to using this
for access, rather than MacX.  We also use the Mac clients for the Oracle
database server via SQL*Net if purely numeric/text database searches are
required.  We get around the IC50>100 etc problem by storing two values,
the % inhibition and the concentration.  For an IC50 of 10 this would
involve storing 50 (%) and 10 (micromolar).  For the IC50>100 example we
would store say 22 (%) or whatever the inhibition was at 100 micromolar,
and 100 (micromolar).

Dr. David A. Winkler                            Voice: 61-3-542-2244
Principal Research Scientist                    Fax:   61-3-543-8160
CSIRO Division of Chemicals and Polymers        CSIRO: http://www.csiro.au
Private Bag 10,Rosebank MDC, Clayton, Australia       http://www.chem.csiro.au

----------------------------------------------------------------------

From: Tom Hagadone <9505051719.A10168' at \`pwinet.upj.com>

Regarding your database question; Here at Upjohn we use a chemical structure
add-in for Microsoft Access derived from our in-house Cousin system to
manage personal, project and departmental chemical databases containing up
to about 10,000 structures and their associated data. The add-in supports
high quality 2-D structure diagrams, fast full structure, substructure, and
similar structure searching, hit list manipulation and most features
expected in a full-function chemical database system. Recently, the add-in
was compared to ISIS/Base by an independent group within Upjohn and judged
to be superior according to the group's judgement criteria.

Last week (4/27) I sent a manuscript titled "Capturing Chemical Structure
Information in a Relational Database System: The Chemical Software Component
Approach" to Bill Milne to be considered for publication in JCICS. As
supplementary material to the paper I have included the complete Cousin
add-in for Access for publication on the ACS Internet server if the paper is
accepted. (...)

Tom Hagadone                         e-mail: trhagado { *at * } pwinet.upj.com
Information Scientist                Phone: (616) 385-7223
Computer-Aided Drug Discovery        Fax:   (616) 385-8488
The Upjohn Company

----------------------------------------------------------------------

From: mws - at - tiac.net (Michael Swartz)

There are two programs that "sort of" meet the requirements you mention:
ChemFinder from Cambridge Scientific Computing and Entropy from SoftShell
International. Both programs have minor database and substructure searching
functionality. ISIS/Base from MDL currently meets your requirements better
than anything else though it is quite expensive. ChemFinder has the
advantage of working with Chem3D though it has many irritating information
management deficiencies.

To get full functioanlity, however, you need a major system. NCI is
currently installing our RS3 system - an ORACLE-based structure search
system - to manage their structures. (...)

Michael Swartz
Marketing Director
Scientific Information Systems
PSI International, Inc.

----------------------------------------------------------------------

From Chemical Design Automation News, January/February 1995, p. 10-11:

"Synopsys Scientific Systems Ltd. has launched its innovative chemical
spreadsheet, Accord, for Microsoft Excel, on the Macintosh and Power
Macintosh platforms. (...)

[It provides] scientists with a full range of chemical facilities to
manage, analyze and search chemical structures and associated data
within Microsoft's popular spreadsheet environment. (...)  [U]sers
can store chemical structures and other chemical objects, such as
substituents, directly in spreadsheet cells with retained 'chemical
awareness'. (...)

Standard chemical editors, such as ChemDraw and ISIS/Draw, can be used
to enter and modify structures or queries.  And chemical data, stored
in database systems such as MACCS and ISIS, can easily be imported
into an Accord-enabled Excel spreadsheet with retained chemical
awareness.  Similarly, individual structures or whole chemical data
tables can be exported in a variety of standard formats, such as SD,
SMD and SMILES. (...)

Accord for Excel is compatible with both Microsoft Excel 4.0 or
higher and is available for the Microsoft Windows, Macintosh and Power
Macintosh platforms at an introductory price of L495 ($695).  A 40%
discount is available to academics.

For further information, contact Dr. Glen Hopkinson or Dr. Paul Hoyle,
at Synopsys Scientific Systems Ltd., (...) ++44-113-245-3339; FAX:
++44-113-243-8733; in the US: (609) 734-7431; in Japan: 81-3-3805-6651;
e-mail: sales(-(at)-)synopsys.co.uk."

======================================================================

------------------------------------------------------------------------
 Marc C. Nicklaus                        Lab. of Medicinal Chemistry
 e-mail: mn1 %! at !% helix.nih.gov               National Cancer Institute, NIH
 Phone:  (301) 402-3111                  Bldg 37, Rm 5B29
 Fax:    (301) 496-5839                  BETHESDA, MD 20892-4255
------------------------------------------------------------------------



Similar Messages
05/04/1995:  Re: CCL-Personal Structure/B
05/04/1995:  Personal Structure/Bioassay/Inventory DB's
08/03/1995:  ACS Chicago - CINF Abstracts    - 29 pages document -
07/26/1995:  2D to 3D strux conv: summary
03/22/1992:  Responses to MACCS alternative query
11/30/1994:        Summary: Available Chemical Databases?
11/08/1995:  Re:  Re: CCL:looking for 3D databases
10/18/1991:  SUM: database & software for organic compounds
03/20/1999:  MDL INTRODUCES CHESHIRE FOR ISIS
12/01/1994:        Available Chemical Databases 2


Raw Message Text