JUL 18 i23

DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
PUBLIC HEALTH SERVICE
NATIONAL INSTITUTES OF HEALTH
BETHESDA, MARYLAND 20014

 

July 13, 1973

RR-00785-01A1

Dr. Joshua Lederberg
Professor of Genetics
Stanford University
Stanford, California 94305

Dear Dr. Lederberg:

The National Advisory Research Resources Council (NARRC) con-
curred with the initial review group recommendation that your
application, RR-00785, be approved with reduced budget.

The following quotes represent their impressions of the weak
points of your application.

"The utility of DENDRAL does not appear to
lie in its contribution to MS within the
foreseeable future ... . Other instru-
mentation using GC/MS and library look-up

or camputer matching of spectra... is
efficient, reliable, far less expensive, and
less involved than the proposed method ....

"Of the total range, three sub-projects of
DENDRAL are neither good chemistry nor good
computer science. Real-time computer control
of MS may be feasible given that the computer
programs have been established, but this
portion of the proposal does not contribute
to the basic investigation. Similarly,
c/3-ywr has such low sensitivity that it
requires 5-6 orders of magnitude more per
sample than MS--and it contributes little

to AI. Finally, the second part of B, the
analysis of urinary metabolites, which is
primarily of clinical interest, would be more
simply done by library look-up of spectra of
known, nommal, and abnormal constituents of
body fluids. Such a campendium has been
compiled by Dr. Markey of the University of
Colorado."
Dr. Joshua Lederberg - Page 2

From a minority opinion favoring disapproval came the following

notions:

I have repeatedly requested enlightenment on
why very large and expensive facilities are
requested and how it may ultimately contribute
to an understanding which would result in
better medical care .... My feelings are
that funding such large projects, of low payoff
probability, could only result in positive harm
to the total biomedical community ....

The reviewers reinforced the shared resource concept by stating:

"In many ways the most significant aspect of the
proposal is that an AT network resource would
provide an excellent demonstration of resource
sharing which would give that concept a maximum
chance of success."

The budgets and rationale recommended are as follows:
BUDGET: SUMEX and DENDRAL:

 

 

 

 

Figure 1
SUMEX
SUMEX -01 -02 -03 -04 ~05
Personnel $220 ,000 231,000+ 242,500" 255 ,000+ 267 , 500+
Equipment 175 ,000 175,000 175,000 175,000 175 ,000
Supplies 20,000 22,000 24,000 26,000 26,000
Cammmnications 28,000 29 ,000 30,000 31,000 32 ,000
Other Expenses
Maintenance 60,000 60 ,000 60 ,000 60 ,000 60,000
Freight 12,000 0 0 0 0
Miscellaneous 3,000 3,000 3,000 3,000 3,000
Travel 5,000 5,000 5,000 5,000 5,000
$533 ,000 525 ,000 539 ,500 555 ,000 568,500
DENDRAL *39 ,000 134,000 128,300 134,700 141,500
TOTAL $572,000 659 ,000 667,800 689 , 700 710,000

*This is only four months for DENDRAL; the other eight months are the end of the

previous grant period.
Dr. Joshua Lederberg - Page 3

SUMEX Rationale:

l.

Personnel: Funds cover the following positions (see
p. 88 of SUMEX proposal for original request):

SUMEX facility head
Network coordinator
Systems programmers
Engineer

Operators

Application programmers
Technicians _
Secretary

Equipment: The equipment is such as to provide
DENDRAL with the capability for Parts A (minus the
cl3-NMR project) and C. Part B, the real-time
control of MS and the use of DENDRAL to analyze
urinary metabolites, was felt not to contribute to
the major objectives of the AI program. The Bare
system plus TIP was increased to $175,000 to
allow for inexpensive terminals and the like.

(Rind£leish)
(Jamtgaard)

FPRENWEWH Ee

 

Supplies: "Camunication" is essentially as
requested. Other supplies are cut by $20,000/year--
deleting part of the engineering and the lab
supplies. This is in line with not supporting

the on-line DENDRAL work or GC/MS clinical

projects.

Travel: Travel is reduced somewhat.

Other es: "Maintenance" is recommended as
requested ond may possibly be somewhat higher

than is really needed. It is assumed that BRB
will monitor these expenses as well as the hard-
ware purchases to ensure economy. Funds for
purchasing computer time "terminal and computer
services" have been deleted. "Other" has been
slightly reduced. "Freight" is as requested.
Dr. Joshua Lederberg - Page 4

 

 

DENDRAL
(Parts A & C only. Personnel projected from first year.)
-01 -02 -03 -04 -05
Personnel $36,800 116,000 122 ,000 128,000 134,500
Equipment 0 12,000 0 0 0
Supplies 700 2,000 2,100 2,200 2,200
Travel 0 1,000 1,000 1,000 1,000
Other 1,500 3,000 3,200 3,500 3,800
Terminals
Telephone
+$39 ,000 134,000 128,300 134,700 141,500

+This is for a four-month period.

DENDRAL Rationale: Completion of Parts A and C should enable you
to exploit fully the DENDRAL work. Figure 2 shows the recommended
total budget. The personnel figure is derived from the first year
request--multiplied by 3 and with 5% increment in each year. The
equipment budget is for a graphics terminal. Two are requested,
one is recommended. The other reductions are trivial except for
travel. Total travel funds as given in the combined budgets seem
adequate for a project of this size.

The issues still at large with this application require personal
interaction to resolve them. The date of August 2, proposed by
Tam Rindfleish, is satisfactory with Dr. Raub and me for a meeting
at Stanford.

Hopefully, we can negotiate SUMEX to our mutual satisfaction and

a fundable conclusion and lay out a pathway for DENDRAL that results
in a MS resource diffusing its gains to its own disciplinary comunity
as well as clinical research.

Sincerely yours,

>
William Roy Bakér, Jr., Ph.D.

Assistant Chief
Biotechnology Resources Branch
Division of Research Resources