The North American Chapter of the International Chemometrics Society

I occasionally exported data in WKS format, loaded the file into a spreadsheet program, sorted the data, and so forth, and then re-read the file back into Pirouette.. The newsletter anno

The North American Chapter of the

International Chemometrics Society

S.D Brown, Course Accreditation

Departmentof Chemistry and

Biochemistry

University of Delaware

Newark, DE 19716, USA

sdb@brahms.UDel.edu

B Vandeginste, Chemometric Abstracts

Unilever Research Laboratory

NAmICS Newsletter #9 September 1994

Pirouette aim high; they

seek to design a

power-ful, yet user-friendly

soft-ware tool that can tackle

the most frequent types

of chemometric

investi-gations Infometrix

achieves this ambitious

goal by judiciously

se-lecting techniques and

then carefully

implement-ing them Much thought

has gone into the

devel-opment of this high

qual-ity product and it shows

clustering and principal

component analysis; the

classification module

offers K-nearest

neighbors and SIMCA;

the calibration module

contains PLS and PCR

routines The underlying

chemometric theory is

solid Necessary options

are available and

From the Editor's Desk

It's a pleasure to bring you the eighth edition of the newsletter of the North American Chapter of the International Chemometrics Society Dave Duewer and Dora Schnur roped me into, er, ah, asked me to guest-edit this issue It's the one you've all been waiting for, yes, the Election Issue! There's quite a line-up of candidates waiting for you to cast your vote, so don't delay in returning your ballot (p 20) All opinions expressed herein are solely those of contributing individuals; their institutions bear none of the blame

Deborah Illman, Guest Editor

In this issue:

Candidate's Statements, 18  Ballot, 20  Miss Prim, 3Seasholtz waxes philosophic, 4  Happy Birthday NAmICS, 5Education, 8  Letters, 9  Vendor Information on List-Serve, 10Calendar, 12  Chemometrics On-Line Conference, 13

2

spreadsheet holds the

data Generating and

managing data subsets

is easy and efficient

can be zoomed to full

screen or resized The

toolbar also provides

the means to spin 3D

plots, magnify 2D

plots, and identify

points in plots

Pirouette provides

what it calls "array

plots" for certain

objects Consider a

SIMCA model

containing three

classes The scores

window will contain

three miniature score plots; each can be successively zoomed

to fill the quadrant by double-clicking

These miniatures can

be surprisingly informative Other miniatures (called multiplots) of raw datashow up to 231 pairwise variable plots(i.e., 22 variables worth); linear correlations are immediately visible even in the reduced form This variety of data views and the advantageous use of color facilitate the tedious, yet necessary,process of examining alarge data set

The installation procedure is automatic In a few cases, some

customization of config.sys and autoexec.bat might be necessary but these matters are spelled out

in the manual The program accesses at most 16 MB of memory and requires

at minimum a 386 computer with 4 MB

of memory and 5 MB

of hard disk space with an EGA or VGA adapter and mouse A math co- processor is strongly

recommended, as is more memory I ran Pirouette on a 386sx (20 MHz, 4 MB, mathco-processor) and a 486dx (66 MHz, 16

MB); the times given below correspond to the flashier hardware unless explicitly noted The program's worksheet can hold up

to 8000 samples or variables with the limit of the combination being determined by the available memory

Extracting 10 principalcomponents from a 75 sample/66 variable data set took less than

15 sec

Pirouette employs data linking in two imaginative ways

First, in SIMCA, PCR, PLS, the number

of model factors is linked to related ob-jects (plots of model-ing power, residuals, predictions, and lever-age, etc) Thus, with aplot of eigenvalues v

number of factors in one window and up to three linked objects in other windows, click-ing on the desired number of factors in the eigenvalue plot triggers an immediate update of the results inthe remaining win-dows The other type

of data linking allows

a user to select a set of samples or vari-ables in one view of the data and see those selections highlighted

sub-in another view

This greatly simplifies the inspection and/or deletion of outliers

They can be highlighted in the residuals plot using a rubberband box, examined in the prediction plot, and then, with a single keystroke, excluded toform a new subset This feature, coupled with the program's computational speed, makes it realistic to investigate and compare many subsets For example, for my 75x66 data matrix, I could delete

a few variables, re-run PCA, and compare theeigenvalue plots in less than 20 sec.Besides its own data format, Pirouette supports ASCII and WKS formats for both input and output I had no problem getting data into the program I

occasionally exported data in WKS format, loaded the file into a spreadsheet program, sorted the data, and so forth, and then re-read the file back into Pirouette Bundled with Pirouette is MasterKey, a utility which translates data files produced by a variety of commercial instruments into Pirouette, ASCII, and WKS formats To export only certain results, the user can choose to save the contents of an active window This is

handy for transferring

discusses the theory

behind the methods

Pirouette is its printing

capabilities For the

record, I have used (or

tried to use) the

following printers: HP

PaintJet (color), HP

LaserJetII, Apple

LaserWriterIINT, and

two other postscript

printers whose names

it is not surprising thatthe process is rather slow on aging hardware; it took my 386sx about 2 min/page for the LaserJetII (I can't give a time for the 486dx because only postscript printers were available on that system.)

It is also distressing that offline devices are not recognized as such

For those working in apostscript

environment, the otherwise fine performance of the software is

compromised by printing difficulties

The program offers theoption of printing to a file or saving TIFF images, which is a fastway around printing out of Pirouette It took less than 15 sec for the 486 tif file-write Perhaps Windows users imagine saving TIFFs,switching into a Windows program thatrecognizes the format, printing via the Print Manager, and switching back to Pirouette Good idea except for the switching part

Pirouette CAN run inside Windows (but only in standard mode and some applications don't like this) but it cannot task switch

Infometrix is aware of the postscript printing problems and is takingsteps to address them

A Windows version ofthe program is due by the end of the year Itsrelease should make both the print speed and printer driver issues moot I found the staff at Infometrix EXTREMELY responsive and knowledgeable My phone calls and email messages were dealt with in a timely and professional fashion

Overall, Pirouette

is a very impressive product It IS expensive (list $4000 with a 40% discount for academic users) but let's face it, good tools are never cheap!

A free, almost fully functional demo is available so interested parties can investigate the program for themselves Since Infometrix will customize a demo containing your own data, this is a no-risk way to get a real feel for Pirouette and see how it works with your system

For further information:

Infometrix, Inc

2200 Sixth Avenue, Suite 833Seattle, WA 98121phone:

206-441-4696fax: 206-441-0841internet:

infomtrx@halcyon.com

Ask Miss Prim

Dear Miss Prim,

My name is Eddie an

my nayburhood is gonna becum a enterprise zone I wanna start a kemmometrics kumpany wit my pals

on da street What shud we do?

Signed,Eddieand

a LatinSqua

Gentle (sic) Reader: Perhaps you should look into a more lucrative business like statistical consulting There are already gangs of unemployed chemometricians roaming the country looking for jobs In fact, there is an

international crisis, with svante or eighty such gangs

throughout the world These people are mean (not average), sum are squared, and many are in analysis

Do a target

transformation on your goals.

[Questions for Miss Prim (Clare Gerlach) may be sent in care of the Editor-in-Chief.]

A rose by any other name…

Mary Beth Seasholtz, mseasholtz@dow.com, (517)636-3646The field of chemometrics is fortunate enough to have progressed to the point where there are multiple generations

of ideas As a graduate student in 1989 eager to learn the tools of the trade, I was confronted with two sets of

‘generations’ of equations describing PCR (and who knows how many for PLS, but that is another story!) This newsletter seemed a good place to present a few lines which demonstrate the equivalence of the two approaches, and to touch on the historical context which led to the move Please forgive the omission to the multitude of appropriate references – Deborah only gave me one page

The older of the two approaches begins with assuming R = TPT where T has orthogonal columns and P has

orthonormal columns In the mid 1960’s when chemometrics was born, there were a few methods available for

calculating T and P One was the not so well behaved NIPALS algorithm Alternatively, T could be obtained by solving for the eigenvalues and eigenvectors of RRT (a square symmetric matrix), and then P could be estimated given R and T The symmetric eigenvalue problem was studied for many years; the most famous book on the

subject was published in 1965 by J.H Wilkinson (The Algebraic Eigenvalue Problem) However, computational

algorithms were not in high demand as computers of the 60’s certainly were not what they are today The

calibration problem then was cT~x (~ indicates truncation) Solving for x via the normal equations gives

c T

r  , giving

c T T T r x t

un un



In 1969 Gene Golub made the singular value decomposition (svd) an algorithmic reality It was long known that an

arbitrary matrix could be written as the product of three matrices, R = USVT, where U = eigenvectors of RRT, V = eigenvectors of RTR and RRT (they are the same) But, until Gene and his coworkers came on the scene there was not a direct calculation (you had to go through the covariance matrices as described above) With the advent of the widespread availability of the svd (and other useful code) through facilities like LINPACK, EISPACK, Numerical Recipes and Matlab, the PCR story has since evolved and a new word is being used: pseudoinverse The calibration

equation now reads c = Rb, and b Rc



ˆ , where R V~S~1U~Tc pseudoinverse Prediction is simply

c U S V r b r

un

T un un

~

~

~ˆ



Equation (2) sure looks different from (1)! Well, recall T could be calculated from an eigenvector problem of RRT

… in fact T = US and P = V Making these substitutions into (1) give c rTV U S U S 1 U S Tc

un

un ~((~~)(~~)) (~~)

which is equation (2), after reduction using standard linear algebra rules

As can be seen, it was because of some relatively new technology in the area of numerical linear algebra which gaverise to the new look for PCR In addition to all the other things that keep us busy, I think we must be as diligent as

we can to continue to bring into chemometrics new developments from disciplines such as applied mathematics, statistics and numerical analysis

See what too much tequila can do?

Happy 20th Birthday to the Chemometrics

Society! It was on June 10, 1974 that the

Laboratory for Chemometrics met with Svante

Wold in Seattle over some great Mexican food

and too much tequila and formed the Society

Our focus was on improving communication

between chemists, statisticians and

mathematicians We also wanted all chemists to

be about 10% chemometricians to insure that

experiments would be designed optimally and all

information would be extracted from chemical

measurements Well, the Mexican restaurant is

no longer in business but the Society and field of

chemometrics is alive and doing very well.

I was reading Chemometrics Society

Newsletter Number 1 (I have a complete set)

published in January, 1976, and it reported 101

members worldwide with half of them owning

the program ARTHUR which some of you may

remember The newsletter announced that the

second FACSS meeting had an attendance of 200

at the "Chemometrics in Analytical Chemistry"

session with papers from Wold, Deming,

Horlick, Duewer and Kowalski Also announced

was the "Chemometrics: Theory and

Applications" session at the Summer 1976 ACS

meeting in San Francisco that later produced the

first book on chemometrics The newsletter

ended with comments, requests and suggestions

from Richard Cramer, Ken Loach and Harold

Martens So much for ancient history.

Two journals, thousands of papers and reviews and dozens of books later we find ourselves today with rich areas of application, powerful chemometrics tools and essentially infinite computer power We are very busy scientists Also, scientists, statisticians, and mathematicians and even chemical engineers have discovered chemometrics and the race is on What will our science be like in the next century, the year 2000?

Allow me to make a few predictions You can use leave-one-out cross-validation to estimate the PRESS, SEP or RHSCV if you choose First and foremost we should all see the necessity to have chemometrics permeate the formal education of all chemists, not only with graduate level courses, training courses and workshops but also

at the beginning levels of chemical education The old "scientific method" that relies on a lot of theory and few definite measurements must die

It should be replaced with equal amounts of theory, experimentation, measurements and simulation and emphasize the multivariate nature

of the world around us There is no place for univariate thinking in our multivariate, dynamic world.

Next, chemometrics will no longer be just a collection of our data analysis methods The tools of chemometrics will spawn new measurement theories that will guide chemists in all areas of research To this end a young

chemometrician, Karl Booksh, and I offer a special report in the August issue of

ANALYTICAL CHEMISTRY titled "Theory of Analytical Chemistry." I invite you to read this paper and incorporate it in your research and education activities I also encourage you to expand this theory and move it into areas of chemistry beyond chemical analysis.

Finally, the tools of chemometrics will move

from mathematics and software to firmware so as

to be transparent to the user and very easy to use

and hard to abuse Our current software, while a

great improvement over ARTHUR and SIMCA,

is still too difficult to use The younger

generation of chemists have good backgrounds in

linear algebra and have little difficulty with

multivariate methods However, the older

generation that will be with us into the next

century doesn’t understand our methods and

therefore prefer to separate one peak from all the

rest or correlate one molecular property at a time

to molecular activity thereby missing the most

important part of nature, covariance We must

accept the responsibility to make it easy for all

chemists to incorporate multivariate methods into

their work Use this as an analogy We are all

expert users of TVs, VCRs, cellular phones and

the like, but how many of us are truly familiar

with the complex subsystems of these devices

To really be of use, our methods must be

integrated into instruments and experiments to

the point of being transparent.

I'll stop here Many of you have your own

vision of the future of chemometrics that I for

one would surely love to hear Perhaps you can

use the newsletter as your vehicle to share your

thoughts with the rest of us and thereby continue

a twenty year tradition.

Q: How many Seattlites does it take to screw in a light bulb?

A: Two One to change the bulb and one to hold both lattes

Q: How many Zen masters does it take to screw in a light bulb?

A: A tree in a golden forest

Q: How many surrealists does it take to change a light bulb?

A: Two: one to hold the giraffe, and the other to fill the bathtub with brightly colored machine tools.Q: How many IBM types does it take to change a light bulb?

A: 100 Ten to do it and 90 to write document number GC7500439-0003 Multitasking Incandescent Source System Facility of which 10% of the pages state only “This page intentionally left blank.”

Q: How many Vulcans does it take to change a light bulb?

A: Approximately 1.00000000000000000000000Q: How many existentialists does it take to screw in

Open a Window on Chemometrics

by

Judith Barnsby

BARNSBYJ@rsc.org

Window on Chemometrics is a new monthly

publication reporting on the latest work in the

computer handling of analytical data It covers the

science of chemometrics and its applications in

spectroscopy, chromatography, and other analytical

techniques

Produced by scanning the international scientific

literature, including the major chemometrics and

analytical chemistry publications, Window on

Chemometrics gives reports of developments in the

following key areas:

General Techniques & Statistics

Calibration & Validation

Computer Programs, Expert Systems and

Applications

Spectrometry

Chromatography

Other Analytical Techniques

Each report gives title, detailed abstract and

Second order analytical instruments or

instrumental methods (i.e those that give a response

matrix when analyzing a pure analyte) have the

advantage of the ability to analyze mixtures which

contain unknown interferences However, this

advantage can be lost, if a suitable calibration method

is not used A medium-rank second-order calibration

method is proposed (full details given), based on

least-squares restricted Tucker models With this

method the second-order advantage is retained

Ordering details:

Window on Chemometrics ISSN 0966-9086

12 issues per year

1994 subscription prices:

US $162.00Canada L95.00 (+GST)

EC & Rest of World L90.00Please order from:

The Royal Society of ChemistryTurpin Distribution Services LtdBlackhorse Road

Letchworth, Herts SG6 1HNUK

Tel: +44 (0)462 672555Fax: +44 (0)462 480947

For further details and a sample copy of Window on Chemometrics, please contact:

Judith BarnsbyThe Royal Society of ChemistryThomas Graham House

Science Park, Milton RoadCambridge CB4 4WFUK

Tel: +44 (0)223 420066(Toll free in US: 1-800-473-9234)Fax: +44 (0)223 423429

E-mail: barnsbyj@rsc.org

Chemometrics is for undergraduates, too

by Nick C Thanasoulias

[Thanasoulias writes to us from the Chemistry

Department at the University of Ioannina, Greece

Ed.]

Many colleagues, who teach in higher education

institutions, keep wondering whether undergraduate

students should be presented with chemometrics or

not Common questions involve the student's

background in mathematics, knowledge of computers

and ability to catch the idea behind the calculations

that chemometrics requires One common argument

is that chemometrics is needed only if you plan to do

research So why not stick to the old familiar

Gaussian least squares?

However, the fact is that everyday we are more

and more confronted with the problem of agreement

between results of different laboratories Most

industrial processes require pilot experiments which

can be performed only by means of applying

statistical and mathematical techniques to chemical

problems Quality control, especially in the clinical

laboratory, is another major application of

chemometrics

It is more than certain that chemometrics can not

be ignored In my opinion it should be taught in the

undergraduate level and students should become

familiar with as many statistical techniques as

possible

In the University of Ioannina, Greece,

chemometrics is taught during the last semester of the

fourth year Since students are able to choose their

classes during that semester, chemometrics teaching

can be modified according to the number of students

attending the class It usually consists of two

modules: one theoretical and one experimental The

theoretical part includes some six to ten hours of

classroom teaching and the chapters covered include:

fundamentals of statistics, significance testing,

analysis of variance, experimental errors, simple and

multiple linear regression, factorial design and

cumulative sum techniques

The second module consists of a series of experimental classes which are designed so that they resemble a "miniature" thesis preparation Each student is asked to apply to a real chemical problem some of the theoretical principles learned so far The outline is as follows:

Definition of the chemical problem;

Literature study related to the chosen chemical problem (usually covers the last 2-5 years and students are asked to concentrate on review papers);

A normality test is applied to a set of repeated measurements in order for the student to decide whether to use parametric or non-parametric techniques We usually take care that parametric statistics can be applied since students are not familiar with non-parametric methods and algorithms;

Classic one-at-a-time factor-change experiments are performed, in which change in an appropriately chosen variable (the response) as a function of one factor is followed while the other factors are kept constant The choice of the factors and response relies on previous knowledge (usually from the literature study) and care is taken not to include more than one factor which may not have

a significant effect on the response The results are treated with the usual regression techniques;One-way ANOVA tests are carried out for each factor

in turn for the student to get an idea of how significant the effect of each factor may be;

Up to four factors are selected in a complete factorial design (five, if the measurements are very easy toperform) The factors are chosen in two levels and all the trials, consisting of all the possible combinations, are carried out twice for an estimation of the effects and the residual variance;

Finally, the student presents a report that follows closely the principles of scientific paper writing and is asked to make conclusions such as the suitability of the system for analytical purposes (limit of detection, limit of determination, sensitivitiy) and to propose ways of maximizing system response

So far, we have tried systems such as the

oxidation of pyrogallol by various oxidants with

chemiluminescence detection, the determination of

gallic acid and tannins, the correlation of analytical

methods for determination of glucomse content of

foodstuffs, and the synthesis of zeolites and their

ion-exchange capability, among others

Letters to the Editor

Communication gap between

QSAR and Physiologial

Modelers What to do?

Dear Editor,

… an increasing percentage of activities

supported by public and private sector organizations

are framed against the complexities associated with

assessing and managing the many forms of risk as

may be posed to humans and the environment by

chemical, biological, and physical agents

A "Workshop on Decision Support

Methodologies for Human Health Risk Assessment of

Toxic Substances" was held November last year

The focus of the Workshop was on the status,

direction, and utility of models described as PBPKPD

(Physiologically Based/ Pharmacokinetic /

Pharmacodynamic means to model metabolic

disposition of chemical substances) and similar

discussions on efforts in QSAR (Quantitative

Structure Activity Relationships)

The Workshop was funded by: Agency for Toxic

Substances And Disease Registry (ATSDR), National

Institute of Environmental Health Sciences (NIEHS),

National Cancer Institute's (NCI) Division of Cancer

Etiology (DCE), Environmental Protection Agency

(EPA), Wright Patterson Air Force Base, Toxicology

Division, and the National Library of Medicine

(NLM)

I had two roles, one as a member of the

Workshop's steering committee and as one of the

wrap-up speakers My presentation focused on the

array of data and information resources needed to

efficiently and effectively support the development,

testing, application, and validation of means to model

the effect of chemical, biological, and physical agents

on biological systems The resources generally useful

in organizational decision making are relevant to assessing and managing the many forms of risk.What was distressing to me and others at the Workshop was the almost complete lack of communication between the PBPKPD and QSAR modelers which raises the challenge as to what should be done to insure that understanding evolves

of the inter-relationships of these models

The steering committee will be maintained, and I will continue as a member I would like to be able to use it as a platform to identify needs for data and information resources which so far have not been identified, and to use this platform to help prioritize forms of such resources to support scientific efforts.Sidney Siegel, Ph.D

Chief, Office of Hazardous Substances Information301-496-5022; FAX 301-480-3537

Expert System Available

An expert system is now available, at no cost, which will determine chemical class, molecular weight and target compound identity from low resolution mass spectra The target compounds are

75 volatile toxic and related compounds Class and

MW information is valid for compounds other than the target set Description is provided in D.R Scott, Anal Chim.Acta, 285, 209-222 (1994), and in forthcoming paper by D.R Scott in Chemometrics and Intell Lab Sys., accepted February 1994 For a copy of the program and instructions, send a 3.5 or 5.25-inch MS-DOS formatted diskette to D.R Scott, AREAL, MD-77, U.S EPA, Research Triangle Park, N.C 27711, USA

News from the President-Elect

Vendor information on List-Serve

Message to List-Serve Readers:

An important role of NAmICS is to provide a

conduit for communication between those who

create and/or apply chemometrics and those who

provide the instrumentation and software to

implement it In the case of software we have

attempted to do this through reviews in our

newsletter.

Unfortunately this approach takes a great deal

of time I spent about 60 hours on the review

that I wrote for the newsletter Although I

personally hope that others will offer reviews of

software and instrumentation that they use, we

have had little success in finding those willing to

do so.

It has been suggested that we allow software

and hardware companies an opportunity to

present their products to our members We, the

officers of NAmICS, agree with this suggestion,

but wish to avoid long sales pitches or

monologues Yet it would be helpful if the

vendors provided the philosophy, scope,

contents, references, prices, and purchasing

procedures for their products It will take some

trial and error and feedback from the

membership in order to define the line between

useful information and annoying advertisements.

There is nothing to prevent any member of

the list-server from posting an advertisement

(other than the VERY real risk of alienating

possible customers) For this reason, we suggest

that presentations first be sent to me for initial

screening and negotiated editing.

Our first attempt at providing information

about a software package will follow this

message Please let me know what you think

about its form and presentation In making these

comments please also remember that you signed

onto this list-serve in order to be kept current in the development and application of

chemometrics.

Donald Dahlberg, Ph.D.

Department of Chemistry Lebanon Valley College Annville, PA 17003-0501 office: (717)867-6143 fax: (717)867-6124 E-Mail: Dahlberg@ACAD.LVC.EDU

Chemometrics Software Upgrade PLS_Toolbox Version 1.4 For Use with MATLAB*

Now Available Submitted by Barry M Wise

***************************************

I am pleased to announce that the PLS_Toolbox Version 1.4 is now available This upgrade of the PLS_Toolbox is the most extensive since it was first released in 1991 The toolbox is now completely compatible with MATLAB 4.x, and takes advantage

of many of the new MATLAB features In addition, I

am now offering technical support for the PLS_Toobox through my home e-mail account and home FAX

Many new functions have been added to the toolbox, and others have been improved considerably.Some examples are:

* Multivariate instrument standardization with additive background correction

* Two and three dimensional scores and loadings plots with labelled points

* Locally weighted regression with y-distance weighting

* K-means statistical cluster analysis with dendrograms

* Savitsky-Golay smoothing and derivatives