Wed Jan 2 13:33:31 1991 1 et yw -2 199]
ST

To: (monica riley) mriley@hoh.mbl.edu

Fee: inbox

Reply-to: (J. Lederberg) lederberg@rockvax.rockefeller.edu
Subject: metabolism data base

Hi, Monica -- Of course I remember you! And very much appreciate
your recent book and Micr. Rev. on The Bacterial Chromosome.

The data base you’re contemplating needs a lot of discussion and
thought to see how it might best complement existing efforts.
(E.G. Mary Berlyn at Yale on E. coli genome; and all the GenBank
etc stuff.) And a molecular biology bibliographic project at
Stanford (Ted Shortliffe, on which I’m a co-PI).

And you should consult Dennis Smith at Molecular Design Ltd.
(Calif) about representations of organic reactions; likewise
look at Index Chemicus (Institute of Scientific Information).

There are some tough conceptual problems defining the boundaries
of "metabolism" and development. Where would you put regulatory
methylation (or any of the rest of the chemistry) of DNA?

I’m not sure you end up with anything less than the entire extant
literature as the database, and a desire for assurance that one
can use metabolism-oriented keys for searching that. If I’m

not mistaken, Medline already uses the EC numbers in the keyword
list but these are likely to be neglected for commonplace
reactions. When we get to full text searchable files we'll be

in shape to use much more sophistication.

So I guess the short answer/bottom line of my views is that you
ought to focus your efforts on systems to aid in extracting
metabolism information from the existing total fount rather than
trying to download a specific "metabolism database". Believe me
there’1l1 be lots to do on that! E.G. the semantic network to
search down pathways of conceptual linkage. Cf. the NLM’s UMLS.

I’m working myself on a thorough analysis of the concepts entailed
in the "central dogma" of

DNA (<)=> RNA ... => polypeptide sequence => folded active protein

and some of the feedback loops. There’s lots to do with just the
dozen central actors in that chain!

If you decide to go ahead anyhow (or in any event) let’s stay in
touch. I have some unorthodox ideas of how a database should
function.

If you’re interested in enzymes, you may be amused at the
following:

 

C) \f£I3) ENZYMES CATALYZE CHEMICAL REACTIONS., viz. are not consumed.\fR.
It is however difficult to find any experimental studies that
would define the rate of "consumption" of enzymes, viz. their
Wed Jan 2 13:33:31 1991 2

thermal stability in presence and absence of their substrates.
It is anecdotally suggested that substrates may stabilize
enzymes in the course of purification (or perhaps enhance
protein folding in the vector expected by the investigator.)

There is one situation where the enzyme consumption is understood:

In fact, experimentally contrived suicide substrates are designed to titrate
the corresponding enzymes, and these are perforce "consumed".
This is a natural function of the "alkylation adaptive" repair
enzyme, O-6-Methylguanine- DNA Methyltransferase, which is
stoichiometrically inactivated by each methyl group it extracts
from target DNA.
cf: Lindahl-T Demple-B Robins-P
Suicide Inactivation of the Escherichia-Coli 0-6-Methylguanine-
DNA Methyltransferase EMBO JOURNAL 1: 1359-1363 1982

Should that be called an enzyme?
Do you know any other natural suicide substrates? (leave out
peroxidation).
Yours ,for an uneventful 1991, Josh
(1990 was too eventful for my taste; and I’m not too hopeful).

P.S. I tried to hire Peter Karp, but he wouldn’t leave Calif.
for long. Mick Noordewier is working with me on computer matters.