Grant Nc. RROO311-o4
Section IV

a

Name: Kaplan, B. Project: PSYCHO
Department: Psychiatry

Project Description: The general nature of the project is the investigation
of the relation of automatic nervous system (ANS) responses to attentional
sets (thinking and environmental observation). In a recently completed
study, which examined ANS changes as a function of verbalization and
environmental attention and rejection, ACME was used to obtain heart rate
and skin conductance values from raw data. This physiological data was
then analyzed, using analysis of variance and other parametric statistical
procedures. In addition, ACME was used for obtaining correlations between
the physiological responses and personality variables.
Future work will look at ANS responses in lactating and non-lactating
human females and will investigate ANS correlates of attention and infor-
mation processing. ACME will be used for transformation of the raw
physiological data into meaningful units and the statistical analysis of
this data,

Name: Kapp, DP. Project: CHBR
Department: Radiology

Project Description: Data analysis of sedimentation patterns of DNA
following X-irradiation,

Name: Kessler, 8. Project: MATSPEED
Department: Psychiatry

Project Description: Will be analyzing mating speed experiments as previously,
including use of linear regression and anelysis of variance programs.

154
Grant No. RROO411-04
Section IV

Name: Kriss, J. Project: BLDVOLL

Department: Nuclear Medicine

Project Description: The program is being used to calculate plasma volume,
blood volume and red cell mass in patients who receive radioactive
tracer material. The determination is useful in the evaluation of
patients with anemia and polycythemia.

Name; Lederberg, J. Project: GENLIBL

Department: Genetics

Project Description: This project contains the statistical and miscellaneous
programs used by the Genetics Department.

Statistical programs: General statistical analysis for the calculations
of sum, mean, standard deviation, the analysis of variance, chisquare and
probability of chisquare distribution, correlation and regression analysis,
the normal distribution with the same mean and standard deviation for
fitting a curve.

Plotting Programs: Plot bar graph in 100 positions, plot of percentage dis-
tribution, plot by function scaled to the range of 0 to 100, plot of
multivalued function allows the choice and supersition of several characters.
Flag is inserted on the chart when underflow or overflow occurred.

Sorting Programs: Sorting a vector in ascending order, sort array and
alphabetical informations.

155
Grant Wo. RROOZ11-04
Section IV

Name: Leiderman, P. Project: PREMIE
Department: Psychiatry

Project Description: Studies of maternal behavior in non-human mammals have
suggested that the degree of interaction permitted between mother and
infant in the postpartum period influences later maternal attachment
and infant development.

Assessments of each mother and her infant are made periodically during
the time the infant is hospitalized and during the first two years at
home. Three major areas are included in these assessments: 1) maternal
attitudes regarding her relationship with the child as reflected in res-
ponses to interviews and questionnaires, 2) maternal behavior observed
during routine caretaking of the infant, and 4) the behavioral deveLlop-
ment of the infant.

Analyses will be done comparing the different groups on various attitu-
dinal and behavioral dimensions at specific points in time as well as
changes within groups over the two year period. We will also examine
the interrelationships between variables as well as the effects of
birth order and sex of the infant on the variables of interest.

156
Grant No. RROO311-04
Section IV

Name: Luetscher, J. Project: Blood-pr
Department: Medicine - Metabolic Research

Project Description: Our research project deals with the secretion and metabolism
of adrenal hormones. Various steroid hormones, catecholamines, and trophic
hormones are measured under different conditions of sodium loading or sodium
depletion. We attempt to define and relate groups of measurements which
assist in the identification of curable forms of hypertension.

The ACME system is used in this project in several ways:

(1) To assist in the calculation of laboratory data, involving
difficult procedures, such as the double isotope derivative method,
which requires solution of simultaneous equations.

(2) Interpretation of data: (a) Simulation of complex systems
(for example, the study of reactions in which substrate and enzyme
concentrations are regulated by different systems and may vary inde-
pendently; or distribution and metabolism of hormones). (b) Statistical
analysis (comparison of mean and variance of different groups, or
correlations between different observations). (c) Analysis of clinical
information (the association between different observations, or the
effects of drugs, diets, etc.).

(3) Data files are used to store information at various stages of
a sequential process, and for collection and analysis of the large
amount of clinical and laboratory data which accumulates during a long
investigation,

(4) The computer has been used extensively in research training

of fellows and staff, first in principles and techniques of computer
use, and subsequently in practical applications.

LT
Grant No. RROO311-04
Section IV

Name: Melges, F. Project: TEMPO

Department: Psychiatry

Project Description: Relating changes in temporal experience to changes in
psychopathological symptoms such as depersonalization and delusions.
Our plan is to discover how aberrations in temporal sequencing and dis-
tinquishing memories from perceptions and expectations relate to the
emergence of definable psychopathological processes, The overall notion
for a number of sub-projects is: confusing past, present, and future
underlies the central symptoms of psychosis.

Sub-projects involve specific attention to drug-induced psychoses and
certain symptoms of psychoses, especially depersonalization, changes
in body image, paranoid delusions, and feelings of influence or alien
control.

Previous work has demonstrated that this approach is highly useful for
understanding psychotic processes. Although aggregate data across sub-
jects will be used, the usual approach will be the intensive study of
Single subjects as they change over time. This allows for the study of
what changes with what over time. In other words, various processes,
i.e. sequential relationships between changes, are discovered by this
intensive approach.

158
Grant No. RROO311-04
Section IV

Name: Miller, R. Project: BIOSTOP
Department: Community & Preventive Medicine

Project Description: The Division of Biostatistics, Department of Community
and Freventive Medicine, provides biostatistical consulting to persons
doing medical research in the Stanford Medical Center and occasionally
elsewhere, such as Presbyterian Hospital in San Francisco.

Biostatistics also trains postdoctoral fellows and some graduate stu-
dents in Statistics in biostatistical consulting. Investigations of

statistical methods may also be done by the staff, postdoctoral fellows,
or graduate students.

The ACME computer is used in all these activities.

Name: Miller, 2. Project: THESES
Department: Community & Preventive Medicine

+

Project Description: This "project" will be used for computing done by or
for graduate students in statistics and biostatistics as part of the
research necessary for their theses, or otherwise connected with their
education.

Name: Miller, R. Project: COURSES
Department: Community & Preventive Medicine

Project Description: This "project" will be used for computing done by or
for the staff in connection with the teaching of courses in biostatistics.

159
Grant No.RROO411-04
Section TV

Name: Nelsen, T. Project: ADRENAL

Department: Surgery

Project Description: Clinical cancer research record protocols and data for
storage and analysis. Next step will be display routines.

Name: Payne, R. Project: SERNAL

Department: Medicine - Hematology

Project Description: The research is concerned with extending and classi-
fication of leukocyte and/or tissue antigens by serologic and genetic

analysis of specific human antisera. In the computer programs, 1)
donor-recipient pairs of individuals are selected for deliberate immuni-
zation to produce antisera, and 2) the antisera are analyzed by comparing
their reactions with test cells in 2 x 2 tables for associations between
them. The significant associations are calculated, and the frequency of
positive reactions are determined.

160
Grant No. RRCOS11-04
section IV

Name: Porter, R. Project: ATC_KIN
Department: Biochemistry

Project Description: Project ATC KIN has been used for data processing in
the investigation of the steady~state kinetics of the enzyme, aspartate
transcarbamylase. The programs were written for curve-fitting the data
from different types of kinetics experiments. For example, program
"DataFit" employs a simple linear least squares fit to calculate the
initial rate of reaction from the raw data, measured as amount of
radioactivity versus time of reaction, and converts the result to
standard units of specific enzyme activity, using units of concentration.

Other programs calculate the kinetic parameters using a linear least
squares fit for the reciprocal transformation, due to Lineweaver and
Burke, of the Michaelis-Menten equation. Another program, "Hyper-

Fit", was written to fit the non-linear, hyperbolic function of the
original Michaelis-Menten equation. The program employs a reiterative
procedure of trial-and-error testing for optimal fit. Trial values

for the two parameters of the Michaelis-Menten equation are tested for
minimizing the residual, and the procedure is reiterated to give a close
approximation of the best values.

This computational procedure was refined and optimized for speed, and
the program will compute the best values for the two parameters, to an
accuracy exceeding that of the data, in less than ten seconds of
computer time in a time-sharing environment. This approach proved to
be so successful that it was adapted for curve-fitting other, more
complex non-linear kinetic equations with more parameters. For example,
the non-linear equation describing substrate inhibition, with three
parameters, which does not give a simple linear reciprocal transformation,
has been employed directly, using this trial-and-error technique. In
addition, the family of hyperbolic curves described by the equation for
competitive inhibition, with three parameters, has been employed
successfully.

The greatest success of this technique has been the use of equations with
four parameters for curve-fitting, such as the equations describing non-
competitive inhibition and parabolic competitive inhibition. With
correspondingly longer times for calculation, these programs have pro-
vided quantitative support for the existence of certain reaction inter-
mediates in the pathway of this enzyme mechanism.

These programs make special use of the on-line communication available

with the ACME system, especially through the use of options and operator-
controlled branchpoints in the programs.

161
Grant No. RROO311-04
Section IV

Name: Reaven, G. Project: DISPLAY
Department: Medicine - Metabolism

Project Description: 1) Derivation of a three compartment model describing
disappearance of plasma insulin. 2) Evaluation of the dog as an experi-
mental model for study of insulin distribution in man. 3) Digital on-line
computer display to investigate the structure of metabolic systems. In-
terested in developing models of glucose, insulin, and triglyceride kine-
tics ag related to diabetes mellitus and atherosclerosis. Clinical data
obtained from tracer studies are analyzed by the ACME computer through
five stages of development. The project DISPLAY includes the third and
fourth stage. The third stage contains the program PEEL that automati-
cally obtains the parameters for a linear sum of constant coefficients
of a system of linear differential equations. The results are used for
the fourth stage. In this phase, the parameters are used for calculation
of a theoretical curve which is displayed on a CRT. It is then compared
with data which also appears on the CRT. The results of PEEL and SKINNER
enable us to change parameters until we obtain a satisfactory visual fit.
In this phase, we used the programs RUNGCURV, DATA, TDATA, berman,
RUNGPLOT, and RUNGFUN. We are now performing similar analysis with res-
pect to triglyceride metabolism and are extending the insulin work. We
expect to use the above procedures the entire year and expect to extend
stage four and develop stage five for analysis of non-linear metabolic
models.

162
Grant No. RROOZ11-o4
Section IV

Name: Reaven, G. Project: FIT
Department: Medicine - Metabolism

Project Description: (See discussion in Project DISPLAY). The final phase
is at the present being developed. The results of simulation in stage
h is to be statistically evaluated to see if the parameters obtained
do not violate the statistical limitations of the data, At the moment,
we are doing this in batch mode at the Computation Center. However,
the most satisfactory results are obtained if the statistical evalua-
tion is coupled to stage 4, We plan to use nonlinear regression
techniques to evaluate these parameters. We have developed two pro-
grams for this procedure; they are RUNGPOW and EXPOPOW, which call
two programs, POWELL and COVARE, which are stored on public file.

Name: Reaven, G. Project: PAT DATA
Department: Medicine - Metabolism

Project Description: (See discussion in Project DISPLAY). This project
includes stage 1 and stage 2. Stage 1 and stage 2 are support pro-
grams for stages 3, 4, and 5. In stage 1, serum endogenous insulin
is biologically assayed by the program INSULIN. The program TCCONRAD
determines triglyceride concentration and the program DOUBLE estimates
the relative weights of data for future statistical analysis. The
second stage consists of statistical support programs for evaluation
of the results of stage 1. The programs used in this phase are:
TEATEST, RAT, CORL, REGRESS, DICK, TG DOSE, RANK, and YOUTEST.

Our research activities also include procedures to determine weight
patterns of patients. The program involved in this aspect is BODYCOMP.

163
Grant No. RROO311-04.
Seeticn IV

Name: Reitan, J. Project: INDIRECT

Department: Anesthesia

Project Description: We are processing cardiac internal timings collected by
non-invasive, indirect techniques in order to monitor the contractile

state of the heart under varying loads and drugs.

Name: Rosenberg, 5S. Project: MEDONCOL

Department: Medicine-Oncology

Project Description: Patient and disease analysis - primarily to correlate
drug responsiveness in cancer patients for educational and research

purposes.

Name: mith, P. Project: PREMIES

Department: Pediatrics

Project Description: This data will be used in any study involving pre-
mature or sick newborn infants cared for in the Stanford nurseries.
Such items as birth weight, gestational age, system disease, etc.,
will be correlated by pairs.

164
Grant No. RROOZ11-04
Section IV

Name: Spevack, A. Project: CONDIT

Department: Psychiatry

Project Description: Classical conditioning experiments of amygdalecto-
myed and intact monkeys.

Name: Stark, G. Project: CHAOS
Department: Biochemistry

Project Description: ACME serves two primary functions. First it is used
to analyze data generated from steady state kinetic experiments on
enzymes, primarily aspartate transcarbamylase from Eschenchia coli.
Second, it is used to process chromatograms generated by an amino
acid analyzer. It is also used for various research and educational
tasks by graduate students and medical students, such as analyzing
the sedimentation velocity of proteins in the ultracentrifuge. Many
of the kinetic experiments referred to above could not be done with-
out the services of ACME, since they involve trial and error fitting
of several parameters to complex functions. The routine processing
of chromatograms on ACME introduces accuracy and dependability not
otherwise available.

Name: Stillman, R. Project: PSYGAME

Department: Psychiatry

Project Description: Our project involves the use of an interactive system
for the interviewing and testing of psychiatric patients.

We are attempting to use the computer and a peripheral CRI display for
anticipated psychological testing. This includes terms from standard
psychological tests and novel tests which the computer is especially
suited for.

165
Grant No. RROO411-04
Section IV

Name: Stocker, B. Project: Stm

Department: Medicine - Micro

Project Description: Main project concerns genetics and physiology of
Salmonella typhimurium - in particular somatic lipopolysaccharide,
flagella and motility, and plasmids, especially R and colicine
factors. I am co-investigator in Dr. Esther Lederberg's project,
which concerns mainly recombination-deficient mutants in the same
organism. Main ACME usage has been 1) in statistical analysis of
data from experiments concerning mouse pathogenicity of lipopoly-
saccharide mutants (estimations of LD50, rate constant for clearance
of bacteria from blood stream, harmonic mean times-to-death, sig-
nificance tests, etc.) and 2) for storage and analysis of extensive
stock-culture data concerning strains with numerous genetic markers.

Name: Strickland, R. Project: GASTRIC

Department: Medicine - G. I, Division

Project Description: I am involved in clinical research projects relating
to gastric secretory function tests and have used the ACME Computer

Facility over the past year for filing of data collected from these
projects and for statistically analyzing the results obtained.

166
Grant No. RROO411-04
Section IV

Name: Summerlin, W. Project: BIOCHEM
Department: Dermatology

Project Deseription: To study biochemical control of collagen formation via
proline/hydroxjproline (ratios).

Name: Vosti, K. Project: VOSTI

Department: Medicine - Infectious Diseases

Project Description: Cross-tabulating variables associated with bacterial
infections such as underlying disease, age, year, prior and contemporen-

eous antibiotics, steroids and immuno-suppressives, appropriateness of
treatment, and outcome (survival, death).

167