SYNERGY » JOURNAL OF UBC SCIENCE

ISSUE 2|2004

2 Words from the Editor

2 Innovations in Mass Spectrometry – Analyzing Ions from Atoms to Proteins

5 Modelling Fluid Dynamics – Weather, Fire and Other Disasters

8 Beating the Odds – Statistical Models Aid Drug Discovery

10 Physiology Under Pressure – UBC Zoologist Walks on the Wild Side

12 Portrait: The Department of Mathematics

12 Projects and Initiatives: Two Grand Micro-Spectroscopy Facilities

14 New Masterminds: Brain Gains at Science

16 Faculty of Science: Kudos and News

UNIVERSITY OF BRITISH COLUMBIA

2 « SYNERGY

Words from the Editor

Chemist Don Douglas and his lab areworking on both sides of the research spec-trum. They are doing fundamental researchon gas phase ions and developing newtechniques in mass spectrometry for ap-plications in biotechnology.

Analytical chemists have been using massspectrometry as a tool for decades. In the pastfifteen years, however, there has been aresurgence of interest in this process for thedetection and measurement of trace sub-stances, in the analysis of proteins and theirbinding mechanisms, and in sequencing smallamounts of peptides.

“The interest of biotechnology and phar-maceutical industries has generated a dramaticgrowth in this area of research,” says UBCprofessor and NSERC-Sciex Industrial ResearchChair holder, Don Douglas, who is working inthe forefront of this burgeoning field.The procedure involves electrically chargingneutral molecules to form ions, which areseparated by electric or magnetic fields

according to their mass and charge. Aquadrupole mass filter is widely used for thisprocess. It consists of four parallel rods withalternating positive and negative rods. By ap-plying direct current (dc) and radiofrequency(rf) electric fields between the rods simul-taneously, ions with a specific mass-to-chargeratio pass through the analyzer and arecollected at the detector, while the rest of theions collide with the rods. Ions with differentmasses are collected sequentially by changingthe voltages, or field strengths, over time. Theoutput, or mass spectrum, is a graph of theintensity of ions versus their mass-to-chargeratio (m/z).

Detecting Trace Elements with ICP-MSMass spectrometry for trace element analysishas become a critical tool in medicine, drugtesting and environmental applications. Typi-cal questions it can answer are: How muchlead is in a child’s blood? How much gold ispresent in a rock sample? What is the level ofmercury in river water?

Douglas helped to develop a technique forthe analysis of trace elements called inductivelycoupled plasma mass spectrometry (ICP-MS).It was first commercialized by Sciex, where hewas principal research scientist before comingto UBC in 1995. In this method, samples areinjected as fine aerosols into argon that hasbeen heated to 5,000 K to form a plasma (ahighly ionized gas in which the number of freeelectrons is approximately equal to the numberof positive ions). At this extreme temperature,the material is vaporized and atomized, usuallyinto atomic ions with a single charge. The ionsare then extracted into a vacuum andseparated according to their mass-to-chargeratios. ICP-MS also can scan for manyelements from a single sample.

“In the old days, lengthy chemicalseparations were required to isolate theseelements,” says Douglas. “Now the separationpart is minimized and the mass spectrometrydoes the cleanup. It is very fast and requiresminimal sample preparation.” Of all multi-element techniques commercially available for

Innovations in Mass SpectrometryAnalyzing Ions from Atoms to Proteins

Dear Reader,

I am pleased to present our second issue ofSYNERGY—Journal of UBC Science. Lastspring, we launched a new look and shiftedthe focus of the publication to give you morein-depth articles on current research anddiscoveries in the Faculty of Science at UBC.Your considerate feedback has been veryencouraging and we have taken yourcomments to heart. For example, you maynotice slight changes to enhance readability.

Again, we are striving to represent thediversity of our researchers and the fields inwhich they work. They are as diverse as ourreadership.

In this edition of SYNERGY you will learnabout chemist Don Douglas’ work in gasphase ions and innovative mass spectrometryinstrumentation for analyzing biomolecules;meteorologist Roland Stull’s research incomputational fluid dynamics to create real-time weather hazard predictions; statisticianWill Welch’s models for high-throughputscreening in drug discovery; and zoologistDavid Jones’ novel method of measuringblood pressure in dead fish.

Our regular features highlight the Facultyof Science’s ongoing efforts in research andeducation. This issue focuses on our Mathdepartment, two major infrastructure grants,and the Faculty's recent recruitment success,

as well as research awards won by the facultymembers in our nine departments.

We hope you enjoy this edition ofSYNERGY—and we always like to hear fromyou! Please e-mail your comments to:synergy.science@ubc.ca

Carola Hibsch-Jetter

SYNERGY » 3

Mass Spectrometry (conventional) Tandem Mass Spectrometry

MOLECULES MOLECULES

MOLECULAR IONS & FRAGMENT IONS

MOLECULAR IONS

IONS OF ONE MASS-TO-CHARGE RATIO

FRAGMENT IONS

IONS OF ONE MASS-TO-CHARGE RATIO

IONS OF ONE MASS-TO-CHARGE RATIO

Ion

Sou

rce

Ion

Det

ecto

r

Ion Detector

Ion Source

Linear Ion TrapTim

e of Flight Ion Filter

Qua

drup

ole

Mas

s Sep

arat

ion

Charging

Mass Separation

Counting

Mass Separation

Fragmentation

Mass Spectrum Mass Spectrum

trace element analysis, ICP-MS has the lowestdetection limits; it can detect elements at muchlower concentrations than other methods.

In a recent project, Douglas developed anew method of extracting ions from argonplasma, which increased the efficiency by afactor of 100—or 10,000 percent—a criticaldevelopment for applications that have smallsamples containing low concentrations of thetarget element.

Innovative Electrode Geometry Whereas ICP-MS produces ions of traceelements with a single charge, electrosprayionization is the technique that Douglas’ labuses for macromolecules—producing proteinions with up to ten or twenty charges, de-pending on the protein.

Linear ion traps use an arrangementsimilar to a quadrupole mass filter to trap,

rather than filter, ions. This is done by placingelectrodes on the ends of the device to createhigh positive potentials to prevent positive ionsfrom escaping. “You can adjust the voltages soions of a broad range of m/z ratios are stablein the trap,” says Douglas. Once all the ionsfrom a sample are trapped, those with a cer-tain mass can be isolated, then fragmentedand separated again by m/z. This process iscalled tandem mass spectrometry.

Douglas and his group are interested inlinear ion traps, versus earlier 3-D ion traps,because the injection efficiency into the trap isnearly 100 percent, instead of 1 to 10 percentfor 3-D traps. Another advantage is that theirvolume can be increased to store more ions bysimply lengthening the rods.

Using tandem mass spectrometry, ions ofpeptides and other small organic moleculescan be fragmented and analyzed for structural

Tandem Mass Spectrometry. Advancingconventional mass spectrometers, Don Douglasand his group analyze complex organiccompounds by employing a second massspectrometer that involves ion trapping and anadditional fragmentation step.

This Linear Ion Trap’s rod set of twodisparate pairs of rods (see below) increases thefragmentation efficiency and the overallsensitivity of the mass analysis.

4 « SYNERGY

information.“If you understand the fragmentspectrum, you can then determine thesequence of amino acids in that peptide,”Douglas explains.

The challenge is getting the ion tofragment before it strikes an electrode. Withthe assistance of Michael Sudakov from theDepartment of Physics at the University ofRyazan in Russia, Douglas and his UBC labhave devised a linear ion trap that favours ionfragmentation by distorting the geometry andadding a second type of electric field (anoctopole field) to the quadrupole field.“Michaeldecided that the best way to add an octopolefield was to make two of the four rods bigger.”Early experiments show that the newgeometry can be at least five to ten times moreefficient at making fragment ions.

“Sciex, our industrial partner, is interested inincreasing fragmentation efficiency, but they alsowant a linear ion trap that can be used as aregular quadrupole mass analyzer,” Douglassays. According to accepted science, however, inorder to do mass analysis, all the rods have to beexactly the same size—within micron tolerances.Undaunted, Douglas and his research assistantChuanfan Ding tried a couple of different rodsets until, unexpectedly, one worked.

“We found that these asymmetrical rodsets will do mass analysis as well or in somecases even better, with higher sensitivity, thana normal rod set, but the key is in how youapply the dc voltage to the rods,” Douglassays. To mass analyze positive ions, they applythe positive dc voltage to the smaller rods. Fornegative ions, the dc polarity is reversed. Hislab is also testing to see if the new rod sets willbe able to do axial ejection—where ions areexpelled out of the linear ion trap in order oftheir mass.

“Other researchers are still writing paperson why 3-D ion traps work better with dis-torted geometry,” says Douglas. “We aretrying to figure out why and how our newgeometry works on linear ion traps.” Hisgroup’s work in innovative instrumentationdesign may be motivated by applied science,but it clearly requires the insight andserendipity of basic science as well.

In order to perform the myriad of biologicalfunctions required by living organisms, thestructures of protein molecules allow them torecognize other molecules in very specific ways,which are critical for binding. But what happensto that protein when you take it out of solutionand put it in a vacuum? Does it “remember” itscomplex folding pattern? Does it bind withother proteins in the same way? Very little isknown about the structure of gas phase ionsand substrates. The other side of Douglas’research is trying to answer these veryfundamental questions.

Studying Gas Phase Biomolecules Electrospray ionization is used to produce gasphase ions at atmospheric pressure directlyfrom ions in solution. This process is gentleenough that proteins up to a molecular weightof at least 100,000 can be ionized. “Somebiologists might say this is a waste of time,because proteins always carry out their cellfunctions in solution,” Douglas says. “But if wecan find conditions where gas phase ionsundergo the same chemistry as they do insolution, then, in principle, we can use massspectrometers to do biochemistry and studysmall amounts of material much more rapidlyand with more specificity.” The applicationsfor screening new drugs and developing “de-signer” proteins would be extremely importantfor Canada’s biotechnology industry.

So far, their work is promising. In a col-laboration with biochemist Grant Mauk (UBCFaculty of Medicine), they found that myo-globin, a protein that binds a heme group(haemoglobin makes blood red), has a similarfolding mechanism in the gas phase as it doesin solution. Douglas is also collaborating withUBC chemist Steve Withers to analyze thebinding of inhibitors to enzymes that his grouphas isolated and studied.

“The hope is that one day we will be ableto screen a protein against a huge library ofcompounds to quickly determine which ones itbinds with,” says Douglas. “It might only workfor certain classes of proteins or smallmolecules, but we don’t know yet; we are stillin the early stages.”

“With funding from both NSERC (NaturalSciences and Engineering Research Council)and industry, we have the freedom to explorefascinating things that may or may not payoff.” – Don Douglas

SYNERGY » 5

Modelling Fluid Dynamics Weather, Fire and Other Disasters

Researchers at UBC’s Geophysical DisasterComputational Fluid Dynamics Centre(GDCFD) study the fluid dynamics ofweather to create multi-scale, real-timeweather hazard predictions, particularly forBC’s rugged mountainous terrain.

The atmosphere is a dynamic fluid of swirlingcyclones, jet streams and turbulent eddiespropelled by changes in wind, temperature,moisture, and pressure. The state of weatherat any one location (over a weather station, atown or a house) depends on how all of thesedifferent variables are interacting and affectingchange in different locations at the same time.Obviously, there is more to weatherforecasting than “meets the eye,” as RolandStull, professor of Atmospheric Sciencesattests.

“Weather forecasters use satellite pictures,radar and surface weather observations, butthese tools only tell them what is currentlyhappening,” says Stull, who is also director of

the GDCFD. “The only tool that can tell us theforecast is numerical weather prediction. If themodel is not perfect, then the forecast won’tbe either.”

Given all of the variables of weather,developing a multi-scale, real-time weatherforecasting model is extremely complex. Tocapture the superimposed motion and scalesof weather, numerical forecasting breaks theatmosphere into grids and then forecasts theaverage condition for each grid cell. Smallergrid cells provide a better resolution of data—and more accurate forecasts, particularly for British Columbia’s mountainous terrain.

“Whenever you average the terrain heightyou are going to miss weather over high peaksand in the low-lying valleys,” says Stull. “Most towns are in valleys, which havedifferent weather than mountain peaks, so weneed to get finer resolution to forecast theweather in steep mountainous terrains.” Finerresolution, however, demands much morecomputing power.

“In order to model the fluid dynamics ofweather, we have to solve for multiplevariables in every location, and how theyinteract with each other at every instant intime, and how this affects the future weatherat all points simultaneously.” – Roland Stull

6 « SYNERGY

“Monster” Computer Tackles Six-Dimensional DataTo produce multi-scale, real-time weatherforecasts, the GDCFD processes over 100gigabytes of data—and 5,000 new animationsand weather maps—per day! “Our data is six-dimensional,” exclaims Stull. “We have thethree dimensions in space and the fourthdimension in time. The fifth dimensioncomprises multiple variables, such as winds,temperatures, and pressures, and the sixthdimension is the multiple realizations we getfrom the different ensemble models.” To handle the huge amount of data, a high-performance computing Linux super-clusterwas built specifically for the centre by IBM.The computer was purchased with funds fromthe Canada Foundation for Innovation, the BCKnowledge Development Fund and UBC.“This research is too expensive to be purelyapplied or purely basic,” says Stull. “Bypooling funds from different funding sources—grants, industry and government clients—wehave been able to afford the computers andthe people to run them.”

Stull and colleagues use different grid sizes toexamine different types of weather events:grids with 108 km cells capture large-scale“synoptic” events such as cyclones; grids with36 km and 12 km cells capture “mesoscale”events such as fronts and mountain circu-lations; and 12 km, 4 km and 2 km cells modelsmaller scale events, such as thunderstorms.

Improving Accuracy with Embedded Grids“We use these nested grids like a multi-stagerocket,” Stull says. “We start with a courseresolution forecast and each subsequent griddrives a finer grid.”

Stull pulls up the website that shows thecentre’s daily, real-time numerical weatherforecasts (http://weather.eos.ubc.ca/wxfcst).At the 4 km grid, we see a mesoscale cyclonemoving east over Ucluelet. We can see theweather pattern is different in Tofino. Stullmoves 24 hours ahead in time and we see thecyclone has reached the mainland. “In themesoscale, we are talking about a circulationthat is only a few kilometres long,” says Stull.“This mesoscale system would probably bemissed by Environment Canada’s forecastingsystem.” In fact, Environment Canada is oneof the GDCFD’s clients.

At the 2 km grid, you can actually forecastdifferent weather in Point Grey, downtownVancouver and North Vancouver. Unfortu-nately, this fine resolution model takes so longto run that even with the centre’s enormouscomputing power (see sidebar) they can onlyafford to make a one-day forecast, versus thetwo-day forecasts for the coarser grids.

Stull and his colleagues at the GDCFD usethe average of four different models of atmos-pheric fluid dynamics to increase the accuracyof their predictions. They have also colla-

borated with various government agencies toconsolidate the different weather networksinto an Emergency Weather Net for BC,which shares data from over 500 locations inthe province. “Wherever there is a weatherstation, we can verify our forecasts every dayagainst observations—and statistically learnfrom our mistakes,” says Stull. The group usesa statistical method called Kalman Filtering toremove bias in the data, in order to determinewhat corrections should be made. Robustmethodology requires verifying daily forecastsover a period of years.

The GDCFD has tailored products forsuch clients as BC Hydro, BC Rail, BCMinistry of Forests, CN Rail, FluxNet Canada,BC Ministry of Transportation and Highways,Parks Canada, and Whistler–Blackcomb.Smaller private companies—an orchard,winery, road service company, and port ter-minal operator—are also part of their diverseclient base.

The centre has received funding from theNatural Sciences and Engineering ResearchCouncil (NSERC), the Canadian ComputerConsortium, Environment Canada, the Cana-dian Foundation for Climate and AtmosphericScience, and the BC Ministry of Water, Landand Air Protection. In addition, le LaboratoireUniversitaire sur le Temps Extrême (LUTE) isfunding the centre to deploy weather stationsaround Greater Vancouver in order to improvelocal forecasts. “We know it is generally rainierin the North Shore than in Tsawwassen, butsoon we will be able to forecast exactly whenand where,” says Stull.

Firestorm ProjectEvery summer, fires destroy large tracts ofCanadian forests. The summer of 2003 was

SYNERGY » 7

one of the most catastrophic fire seasons inBritish Columbia’s—and Canada’s—history.The images were terrifying. Walls of flamesravaged through tinder-dry forests at over 7 km per hour (150 square metres in fourminutes), leaping several kilometres overhighways, waterways and firebreaks, anddestroying everything in their path. Nearly2,500 fires burned more than 265,000hectares at a cost of $375 million. More than30,000 people were evacuated from theirhomes. Ten thousand firefighters and supportpersonnel risked their lives to battle theflames.

These large fires actually alter weather, soforecasting their behaviour is extremelydifficult. The Firestorm Project, led by RolandStull and colleague Terry Clark, was initiallyfunded by Forest Renewal BC in 2001–02,and covered three main aspects of forest fireforecasting. The first involved tailoring real-time weather forecasts to meet the specificneeds of fire forecasting. In the second aspectof the project, Clark used ultra-fine resolutionnumerical simulation of real fires to betterunderstand fire behaviour. “We would usethese observations to try and determine howwell we were able to predict the unknowns,such as fuel loading, burn rate, and fire frontspread,” says Clark.

The name “Firestorm” comes from theunpredictability of these huge fires, which havetremendous heat output and updrafts that suckenormous volumes of air back into the fire.“This fans fire even further, and causes theflame front to move in a direction and speedcompletely different from what forecasterswould predict from synoptic-scale, ambientbackground weather,” says Stull. “So the goalis to model the two-way interaction between

the fire and the weather itself, including theeffects of the steep topography that we havein BC.”

To understand this two-way interaction,Clark’s model examines the heat exchangebetween the fire and the atmosphere. Thismeans working with flamelets and combustionat the molecular level, which requires muchsmaller grid cells, in the range of one to threemetres. “In a moving wall of flames, probablyonly 10 percent is actual hot flame; the rest isair,” notes Clark. “We want to know how theatmosphere arranges itself to exchangeoxygen with the fuel.”

The third aspect of their research involvedflying over forest fires to study the flame-frontdynamics. “Larry Radke was the airbornemission scientist who directed the pilot to fly usover the fires, and Jenn Rosberg was thestudent who was strapped in the back of theplane with an infrared camera and laptopcomputer,” notes Stull. “Circling the fires inthese steep mountains with all of the smoke,cold air and turbulence was very exciting.”

Last year’s fierce forest fire seasonimpelled Forestry Innovation InvestmentLtd.—the BC government’s investmentmechanism for promoting sustainable forestmanagement—to extend Firestorm fundingfor 2003. However, the focus was on real-time weather forecasting rather than fire-behaviour simulation. Even with the GDCFD’senormous computing power, these models arestill too large to run in real time. Stull isoptimistic that with increases in computingpower in the next five years—and crucial newfunding—they will be able to use this basicresearch to better predict and fight forest fires,and help save lives, communities andindustries across Canada.

This July over 600,000 hectares of forest wereburning in the Yukon. British Columbia had1,000 fires burning over 150,000 hectares. (Source: National Resources Canada, 2004;Photos: Phil Maranda)

“What makes the Firestorm Project unique isthat this model also allows the fire to alter theweather, not just the weather to alter the fire.”– Roland Stull

8 « SYNERGY

Beating the OddsStatistical Models Aid Drug Discovery

New Department Head of Statistics, WillWelch, uses statistical models to increasethe efficiency and reduce the cost of high-throughput screening in drug discovery.

Most pharmaceutical research companies havecorporate libraries of hundreds of thousands tomillions of chemical compounds—huge datasets that potentially contain biologicallyeffective substances useful for creating newdrugs. With the advent of high-throughputscreening (HTS) in pharmaceutical research,many compounds can be quickly tested foractivity against a biological target. Although thecost per bioassay is relatively low, approximately50 cents to five dollars per compound, screeningevery compound against every target is expensiveand unfeasible, says Will Welch, new departmenthead and professor of Statistics at UBC.

While HTS allows scientists to assay amillion compounds in a couple of weeks, withoutstatistical tools the error rate can be as high as 10to 40 percent, with a “hit rate” as low as twopercent for finding active compounds. “Even ifyou screen everything, the assays are notaccurate,” Welch states. “The molecules that areassayed to be positive will in many cases turn outto be negative and vice versa.”

One of the reasons for the high error rate isin the methodology. In current high-throughputmethods, speed is attained at the expense ofaccuracy. A typical testing plate used in HTS(roughly 3” x 4”) contains 1,584 wells, eachabout the size of the head of a pin. One targetcompound and one chemical compound aremixed in solution in each well and the amount ofactivity, such as binding, is measured. Withsamples this small, it is easy for errors to occur.

Streamlining High-Throughput Screening withSequential Screening Instead of attempting to test millions ofcompounds, Welch approaches the problem bybuilding a statistical model to predict biologicalactivity based on a small subset of the entirecompound library. With funding from theNatural Sciences and Engineering ResearchCouncil (NSERC), the Mathematics ofInformation Technology and Complex Systems

(MITACS), and industry partner GlaxoSmith-Klein, Welch and his group are honingstatistical methods for sequential screening,which rank unassayed compounds in the orderof their probability of activity—from most activeto least active. By bioassaying only the mostpromising compounds, the “hit rate” for findingactive compounds is increased. Moreover, doingfewer assays means testing can be more care-fully conducted, reducing the error rate. It is aniterative process; whenever additional com-pounds are selected for screening and newbioassay results added to the initial data set, themodel is improved. In one example, screeningless than 20 percent of a collection yieldednearly 75 percent of the active compounds.

Welch uses an analogy of credit card offersin direct mail. Most people throw the mailingaway, thereby falling into the inactive category.Those who apply would be actives. “The creditcard company builds a model of who is likely torespond and only mails to those people,” hesays. “It’s exactly the same problem.”

Tackling Descriptor Variables, SystematicErrors and Psychological HurdlesTo build the statistical model, the three-dimensional structure and electrostatic activityof a chemical compound must be translatedinto quantitative variables. The problem is thatthere are no standard definitions—differentchemists have completely different sets ofdescriptor variables. These might consist of alarge number of simple variables, such as howmany times a carbon is connected to anothercarbon a certain number of atoms away. Orthey might attempt a mathematical descriptionof the properties of a whole molecule.

“Trying to benchmark the differentdescriptors against each other is ongoingwork,” says Welch. “We really don’t knowwhich descriptor set is the best, or whichstatistical method is the best. The question maynot be as simple as what is best, because it maydepend on the type of assay you are doing.”

Systematic errors in the bioassays pose achallenge as well, such as calibration and higherevaporation at the edge of the plate. Whenmeasurements are replicated to check for

errors, the robotic equipment is often set up toput the same compound in the same well onevery plate. “This makes it difficult to tell if ameasurement is a result of a well effect or acompound effect,” says Welch.

There is also a psychological bias favouringdiversity over replication in choosing activecompounds. Given a choice between screeninga large, diverse number of compounds versusassaying a smaller number twice, most research-ers favour diversity. “There is a psychologicalbarrier that we don’t totally understand,” Welchnotes. People believe that they are less likely tomiss something if they assay a larger number ofcompounds against a target, even in thepresence of considerable error.”

Averaging Nearest Neighbours andRegression TreesDetermining how to select a smallerrepresentative set of molecules for sequentialscreening is a major aspect of Welch’s work.The task is relatively simple if some activecompounds have already been identified, ifthere is a crystal structure of the target, or if astructure–activity relationship (SAR), is known.However, when a target is not yet identified—asin many protein–protein interactions—the taskis much more problematic. How do you choosea diverse subset in a descriptor space of severalhundred dimensions in the absence of targetinformation? How do you build a model withhighly proprietary corporate databases, whereyou are working with disguised compounds?

Using public-domain data from the NationalCancer Institute, Welch and colleagues haveadapted the “K-Nearest Neighbour” statisticalmodel, which predicts the activity of anunassayed compound relative to the activity of its“nearest neighbours” in chemical descriptorspace. “Basically it is guilt by association,” saysWelch. “If the compound of unknown activity isin a neighbourhood with many actives, then Ipredict it to be active. For example, if it is neartwo compounds that are active and five that arenot, then we estimate it has a probability of twoout of seven to be active.”

Large molecular data sets contain manyweak or “junk” variables. For drug discovery,

SYNERGY » 9

often only a subset of descriptor variables isrelevant to biological activity. A techniquecalled subset averaging is used to filter theplethora of data and screen out junk variables.

Welch’s group also applied this method toclassification trees, another statistical methodthey are working on. The average ofregression trees based on various subsets ofvariables detected 37 percent more activecompounds than a single tree based on allvariables. “It turns out that if we average acrossall of these models, where some might be goodand others not so good, the performance isclose to that of the good models,” notes Welch.“This is a relatively new finding that I am

working on with collaborator Professor HughChipman at Acadia University and graduatestudent Marcia Wang.” Welch’s other principalcollaborators in drug discovery are RaymondLam and Stan Young at GlaxoSmithKline andRuben Zamar, Statistics professor at UBC.

Sequential screening for drug discoverycan be used to rank virtual compounds as well.“The compounds being ranked could be therest of your corporate sample, or they could bemolecules that have never been synthesized,”notes Welch. “Now we can assay compoundsthat only exist on the computer.” He and otherstatisticians are showing that life in silico canbe multi-dimensional indeed.

Sequential Screening—Statistical “Filter”Approach for Efficient Drug DiscoveryInstead of screening millions of compounds foractivity by time-consuming and costly bioassays,a subset of 1,000’s of compounds is used fromthe data set. Both the chemical descriptors andthe results of the bioassay are incorporated tobuild a model. Virtual (computational) screeningbased on this model immensely reduces thenumber of compounds that need to be testedfurther for activity and at the same timeincreases the percentage of active compounds.

“We believe that not much is learned bymaking a million measurements versus tenthousand carefully chosen measurements.”– Will Welch

Predict Activity as a Function

of x1,...xy

Promising Compounds

Millions of Compounds

Subset of 1,000’s

In the Computer Lab

Computation of Chemical Descriptors x1,...xy of Compound Structures

In the Bio Lab

Bioassay: Screening of the Biochemical Activity

ModelBuilding

Virtual Screening

Org

anic

Che

mic

al C

ompou

nds

10 « SYNERGY

Physiology under PressureUBC Zoologist Walks on the Wild Side

Celebrated zoologist and Order of Canadamember David Jones has pioneered thestudy of diving response and circulatorysystems of a variety of animals. Now, after35 years, his research path has taken asurprising new twist.

The animal care facility where David Jonesconducts his research is hidden away in aremote area of UBC’s campus. It is populatedwith rare and common species from all over theglobe: birds, fishes, giant sea turtles, crocodiles,Australian emus. As we walk past the turtletanks, an enormous 130-pound turtle comes tothe edge and swats the water. “They take greatdelight in splashing you,” Jones laughs—something he does heartily and readily.

Delving into Diving ResponseJones’ basic curiosity and passion for researchhas made him a world leader in his field. In hiswork on comparative physiology of divinganimals, he and former student Russ Andrewspioneered the use of telemetric data loggingdevices to monitor dive depths and times,swimming speed, heart rate, body temperature,and feeding in free-ranging elephant seals andleatherback turtles—work that garnered theFlavelle Medal from the Royal Society ofCanada.

“What you find in really accomplished diversis small lungs, which seems counterintuitive, butin fact they are not using their lungs at all,” hesays. It takes 5 to 10 minutes for a seal to getdown to a depth of between 300 to 500metres, while their lungs have collapsed 30seconds into the dive. Seals use oxygen that isstored in the blood, and when they dive, theirheart rate drops to about one-fifth of thenormal rate, he explains.

Studying Circulatory Structure and FunctionThe second major area of Jones’ research isthe functional morphology of circulatorysystems in both vertebrates and invertebrates.His study of extra cardiac chambers in fishesand frogs, and multiple hearts in invertebratesled to seminal research in the cardiac dynamicsof crocodilians. Unlike all other reptiles,

crocodiles have a completely divided ventricle,which recirculates venous blood to the bodywhile bypassing the lungs. The crocodiles’unique circulatory system is in part whatmakes them such deadly predators, because itallows them to swim silently underwater forhours at a time without surfacing.

One of Jones’ ongoing interests has beenthe bulbus arteriosus in teleosts, or bony fish.It is an elastic chamber that sits between theventricle and ventral aorta. Two basic questionshe has been studying over the past twentyyears are: What is the function of the bulbus inregulating the blood pressure of fish, and doesmorphology affect function? “The interestingthing is that the bulbus comes in all shapes andsizes,” says Jones. “For example, the bulbus ofa marlin is long and thin, and that of a tuna isconical, but they all do the same job.”

In one study, he and colleagues recordedarterial blood pressure and simultaneouschanges in the volume of the bulbus ofanaesthetized yellowfin tuna. When the bulbuswas nearly empty, a small change in bloodvolume resulted in a large increase in bloodpressure. Once systolic pressure was reached(the pressure when the heart contracts) largechanges in volume resulted in very smallchanges in pressure, indicating that the bulbusis responsible for maintaining ventral aorticpressures over a large range of volumespumped by the heart.

Necrophysiology—Studying the Hearts ofDead FishAfter twenty years of research in bulbarmorphology and function, Jones discoveredthat the hearts of dead fish can reveal bloodpressure as accurately as testing live,anaesthetized fish. It was an unexpectedbreakthrough. “There are over 25,000species of bony fishes, so this allows us torecord the blood pressure of a fair number ina way that is quick, infallible and cheap,” hesays. Setting up to measure blood pressure oflive fish can cost in the range of $200 to $300per fish. Tiny fish such as zebrafish, whichweigh less than a gram, cost nearly $20,000because of the specialized equipment required.

The crocodiles’ unique circulatory systemallows them to swim silently underwater forhours at a time without surfacing. (Photo:Manuela Gardner)

SYNERGY » 11

The beauty of the procedure is in itssimplicity. With saline, a syringe and sometubing, it can be done almost anywhere. Asmall amount of saline is injected into one endof the bulbus and then rises up a vertical tubetied into the other end. The height of the fluidin the tube is a measure of the pressure in thebulbus. “If you plot the pressure versus thevolume you are putting in, there is a largepressure increase with the first injection ofsaline,” says Jones. After it reaches normalblood pressure, the plot is absolutely flat.Additional injections further inflate the bulbus,but do not change the pressure. This methodworks for any fish above 100 grams but withmore sophisticated (and expensive) technologyJones has measured pressure in fish as smallas one-third of a gram! To test the method, heplotted the relationship between bloodpressures measured in both living and deadfish. The regression coefficient was a near-perfect .96. “It was the best R2 I’ve ever hadin my life,” he exclaims.

To get accurate measurements, thetesting must take place within 24 hours afterthe fish has died—preferably within 12 hoursfor large predators. However, Jones’ labdiscovered that putting the fish hearts in 30-percent alcohol could prolong the study periodby several days. Most recently, he and studentKevina Perbhoo have managed to makerecordings from rainbow trout that had beenpreserved in the Zoology Fish Museum for 55years! They were still able to obtain accuratemeasurements of that species’ blood pressure.“It really is an amazing organ,” he says withcharacteristic enthusiasm.

Protecting Brains and BrawnJones also discovered a correlation betweenspecies activity level and blood pressure. Toppredators like tuna and marlin have thehighest blood pressures: 100 millimetres ofmercury (133 hectopascal), close to that ofmammals. Lazier species such as perch andseahorses have blood pressures in the range of10 millimetres of mercury (13 hectopascal),and the not-so-ace predators such as trout andsalmon have middling blood pressures.

Jones is anxious to study the bonitobecause it is closely related to tuna and marlin.If the bonito also has a high blood pressure,then this would be characteristic of the group,he explains. However, if the bonito has a lowblood pressure, then high blood pressurecorrelates with heating parts of the body.“Tuna keep their muscles warm and marlinkeep their brains and eyes warm because theyboth go down to really cold water,” notesJones. “This would indicate that you need ahigh blood pressure to develop these regionsof heating.”

This research can provide insight into theevolution of blood pressure in fish, the long-term effects of non-lethal cardiac mutations inzebrafish, and blood pressure as a factor in thepopulation distribution or health of a group. Italso presents a new opportunity to study theeffects of size on blood pressure. “In rats forexample, an adult is only about 20 times thesize of a baby. Whereas a 100-kilogram fishhatches from a tiny egg—so you can get amillion times the size range.”

For Jones, it all comes down to asking theright questions. “If, as a zoologist, you orientyour research to questions of a medical nature,often they are not very interesting becausethey don’t really address the issue of what theanimals are designed to do.”

“Many people collect a great amount of dataand then decide what to do with it. What youreally need to do is to start with an interestingquestion.” – David Jones

Studying diving animals under water: Sincetheir lungs have collapsed after half a minuteinto a 500-metre dive, seals use oxygen that isstored in the blood.

12 « SYNERGY

Portrait: The Department of Mathematics

The Department of Mathematics is knownthroughout the world as a leading centre ofmathematical research. In the past five years,our faculty have won half of the majorCanadian Mathematical Society researchprizes. Recently, Math faculty have receivedmany other prestigious awards, includingseveral Sloan Fellowships and research prizesin Probability and Number Theory.

The department has played a critical rolein the development of vibrant institutes. ThePacific Institute for the Mathematical Sciences(PIMS), centred at UBC, has been a majorforce in the mathematics of North Americaand beyond. Under leadership from PIMS,the Banff International Research Station(BIRS), which hosts weekly workshops inmathematics, was created last year.Mathematics of Information Technology andComplex Systems (MITACS) providesexceptional opportunities for collaborationwith industry.

Currently, the department has roughly 60 full-time faculty, 85 graduate students and25 postdoctoral fellows. Research interestscover a broad array of areas, pure and app-lied. In the past four years, the departmenthas undergone a remarkable period of re-newal with the recruitment of many newfaculty members, including several CanadaResearch Chairs.

The department offers a lively programof courses, seminars, workshops, and infor-mal activities that encourages the study ofmathematics on all levels. This includes manyinnovations to serve the changing needs ofmath students and those in other depart-ments, as well as advanced graduate-levelcourses taught in the summer in conjunctionwith special programs sponsored by PIMSand other institutes.

Through the Institute for Applied Mathe-matics (IAM) and several joint appointments,the department enjoys strong connections

with other UBC departments. Each week,the department, IAM and PIMS host a largenumber of specialized seminars and colloquia,featuring talks by leaders in pure and appliedmathematics from UBC and around the world.

The Math Club provides many study andsocial opportunities for undergraduates inMath and related fields. The departmentoffers weekly training sessions (with pizza!)for the challenging Putnam Exam; ourPutnam team regularly finishes within the toptwenty in North America and often within thetop ten.

Students from all around the worldchoose to undertake graduate study in Mathe-matics at UBC, leading to MSc and PhDdegrees. Math graduates have won manydistinguished awards and have obtained highquality jobs in universities and industry.For more information, please take some timeto browse through the department website:www.math.ubc.ca

Projects and Initiatives: Two Grand Micro-Spectroscopy Facilities

This year Faculty of Science researchers haveearned two significant infrastructure grants infederal funding—more than $8 million forgroundbreaking laser and X-ray spectroscopyresearch from the Canada Foundation forInnovation (CFI). "Our researchers are to becongratulated for their determination tostrengthen UBC research through cutting-edgeequipment and facilities,"said Indira Samara-sekera, UBC's vice-president, Research.

Laboratory of Advanced Spectroscopy and Imaging ResearchProject leader John Hepburn, professor in thedepartments of Chemistry and Physics &Astronomy, and dean of the Faculty ofScience, got good news for his collaborators atUBC and Simon Fraser University (SFU):Their joint application for a new Laboratory ofAdvanced Spectroscopy and ImagingResearch (LASIR) has been granted $5 millionby the CFI. The principal researchers involvedat UBC are the chemistry professors Allan

Bertram, Michael Blades, Michael Fryzuk,Mark MacLachlan, Moshe Shapiro, andMichael Wolf.

LASIR will encompass two hubs with atotal of 5,000 square feet of laboratory andclean rooms within renovated space in thechemistry building at UBC and in newly con-structed space at SFU. Together they will beequipped with seventeen end stations, whichwill be linked to a core of seven state-of-the-artlaser systems. The lasers will span a range ofwavelengths from X-ray to infrared, a varietyof power intensities, and time scales fromnanosecond to femtosecond.

This research facility will help explore newfrontiers in laser–matter interaction. The keyareas of LASIR are laser chemistry and spec-troscopy, new materials and catalysis research,and environmental science. The wide array ofvarious lasers and their innovatively flexiblelaser system to end-station configuration willfacilitate a photolithography centre and spurnew approaches to materials characterization

and atmospheric aerosol research. The great and still growing consequences

of aerosol pollutants have long been neglected.Now environmental scientists will be able toscrutinize the chemistry of such airborneparticles in order to quantify their secondaryhealth effects. Materials scientists will developnew functional materials and processes formolecular-scale semiconductor circuitry andnano-scale devices, and will be able to char-acterize a variety of optical properties of newmaterials. Catalysis researchers will see first-hand the photo-physics of catalytic reactionsand enzyme activity with the time resolutionand wavelength tunability required to unravelthe intricate molecular motions driving thesecomplex systems. Laser chemists and spec-troscopists will have the tools needed todevelop and apply new, high-order nonlinearoptical spectroscopies and microscopies.Using the quantum properties of matter,methods of controlling chemical processes canbe developed. These techniques will help to

SYNERGY » 13

better understand surface processes that areimportant in industry, atmospheric scienceand cell biochemistry.

LASIR will allow researchers—for thefirst time—to tackle fundamental scientificquestions beyond the reach of previouslyavailable instrumentation. Further benefitsanticipated from LASIR include newtechnologies and processes to controlchemical reactions with light and to producenanowires that use light to carry data.

LASIR will be accessible for specialistsand non-specialists alike. Experts from avariety of disciplines, such as physics,engineering, biology, medicine, microbiologyand chemistry, will work side by side, creatingan environment of enriched training forgraduate students and an enhanced exchangeof knowledge. Synergies generated betweenacademia and the private sector will stimulategrowth in advanced technology.

Resonant Elastic and Inelastic Soft X-rayScattering BeamlineProject leader George Sawatzky, professor inthe departments of Physics & Astronomy andChemistry, and his UBC colleagues, physicsprofessors Douglas Bonn, Andrea Damascelli,and Thomas Tiedje, also got their share fromthe CFI Innovation Fund this year—morethan $3 million to create a uniqueexperimental facility involving a novel soft X-ray scattering beamline and end station.This multi-university resource will enable theresearchers to apply spectroscopic techniquesthat have been developed for visible light atsoft X-ray wavelengths. The ensuing highresolution and extremely high sensitivityspectroscopy will make it possible to analyze

chemical elements and magnetic phenomenato view ultra-small structures of materials.

Part of the facility will be built at UBCand then installed at the Canadian LightSource (CLS, University of Saskatchewan).Wavelengths matching the size ofnanostructures will facilitate high-precisionspectroscopy in dilute systems such as singlemagnetic atoms in large macromolecules. Forthe first time, phases in materials that differonly by their electronic or magnetic structureon a nanometre scale will be made visiblewithout electron damage. Atomic site-specificstudies of complex organic systems likepolymers, biological macromolecules, andcells in aqueous environments will be possible.Researchers will be able to determine spatialdistribution of elements or particularmolecular groups and their changes with timeand chemical environment. They will also beable to study important aspects of theelectronic structures in high temperaturesuper-conductors, monolayers of moleculeson surfaces, and artificial structures.

This project will exceed the internationalstate of the art of soft X-ray based researchwhen added to the current end station andcomplementary beamline at CLS. The facilitywill be used not only for pioneering researchbut also for studies into applicabletechnologies. For instance, the proposedinvestigation of magnetism in thin films hasthe potential of evolving a new type ofmemory electronics in computers. The uniqueanalytical capabilities of the X-rayspectroscopy and scattering techniques arealso expected to be exploited by collaboratorsin the environmental consulting industry.

Soft X-ray Scattering Beamline and EndStation. This experimental station is ‘just’ aprototype of what George Sawatzky andcollaborators are going to build at UBC—anew state-of-the-art facility that then will bestationed at the CLS where a new beamline isbeing built according to their specifications.The new facility will be an immense improve-ment and will be much more flexible due to acomplete control of the polarization and high-energy resolution.

14 « SYNERGY

Whether Canadian Research Chair (CRC),cross appointment, junior or senior position—the Faculty of Science welcomesthe new faculty members in the nineScience departments.

Leticia Avilés, Assoc. Prof., Dept. ofZoology; Licenciate Biological Sciences,Pontificia Universidad Católica del Ecuador,Quito; MSc and PhD Organismic andEvolutionary Biology, Harvard University,Cambridge, MA, USA. Research:Evolutionary ecology. Using both empiricaland theoretical approaches, I am currentlyinvestigating the role of multi-level selectionon the evolution of sociality and of life historytraits and local population dynamics inmetapopulations. Empirical studies, primarilyin Ecuador, focus on spiders and otherarthropods.www.zoology.ubc.ca/zoology/z/aviles

Johannes V. Barth, Assoc. Prof., Depts. ofChemistry and Physics & Astronomy; CRC;Diploma (MSc) Physics, LMU München,Germany; Dr. rer. nat. (PhD) PhysicalChemistry, Freie Universität Berlin,Germany. Research: My research contributesto the advancement of supramolecularscience and methodologies for nanoscalecontrol of matter. Notably, we focus onfunctionality and atomistic understanding oflow-dimensional molecular nanosystemsengineered at surfaces, using devisedassembly protocols.www.chem.ubc.ca/personnel/faculty/barth

Giuseppe Carenini, Assist. Prof., Dept. ofComputer Science; BSc and MSc ComputerScience, U of Milan, Italy; PhD IntelligentSystems, U of Pittsburgh, PA, USA.Research: Within the fields of artificialintelligence and human–computer

interaction, I integrate natural languageprocessing and information visualization toultimately generate highly efficient interactivecomputer systems. I study preferenceelicitation, information summarization andthe generation of interactive multimediaarguments.www.cs.ubc.ca/people/profiles/carenini.html

Eric Cytrynbaum, Assist. Prof., Dept. ofMathematics; BSc Mathematics, McGillUniversity, Montreal, Canada; MS and PhDMathematics, U of Utah, Salt Lake City,USA. Research: I apply mathematical andcomputational techniques to problems bothin cell biology and electrophysiology,focussing on the role of the cytoskeleton andmolecular motors in mitosis and on thedynamics of cardiac tissue, particularly in thecontext of cardiac arrhythmias.www.math.ubc.ca/~cytryn

Erik Eberhardt, Assist. Prof., Dept. of Earth& Ocean Sciences; BEng GeologicalEngineering, MSc and PhDGeotechnical/Rock Engineering, U ofSaskatchewan, Saskatoon, Canada.Research: I am interested in geotechnicalrock engineering problems. Specifically, Iapply advanced numerical methods combinedwith geotechnical instrumentation toinvestigate rock slope stability andunderground excavation problems.www.eos.ubc.ca/public/people/faculty/E.Eberhardt.html

Ivar Ekeland, Prof., Depts. of Mathematicsand Economics; CRC; BSc and MScMathematics, Université Paris-Sorbonne,France; PhD Mathematics, Université Paris-6, France. Research: My present field ofinterest is the mathematical modelling ofhuman behaviour. My earlier work was in

geometry and the calculus of variations.

James J. Feng, Assoc. Prof., Depts. ofMathematics and Chemical & BiologicalEngineering; CRC; BSc and MSc Mechanics,Peking University, Beijing, China; PhD FluidMechanics, U of Minnesota, Minneapolis,USA. Research: I study the flow of complexfluids, a group of microstructured materialswith wide applications. I use mathematicaland numerical modelling to explore howinterfaces between immiscible fluids evolveduring flow and processing.www.math.ubc.ca/~jfeng

Sean W. Graham, Assist. Prof., Dept. ofBotany and UBC Centre for Plant Research;BSc Genetics, U of St. Andrews, Scotland;PhD Botany, U of Toronto, Canada.Research: The reconstruction of the Tree ofLife is a central goal of modern biology. Myresearch focuses on making robust inferencesof the major details of land–plant phylogeny,applying these to studies of morphologicaland molecular evolution, and investigatingthe biodiversity of understudied plantlineages. www.botany.ubc.ca/graham.htm

David J. Jones, Assist. Prof., Dept. ofPhysics & Astronomy; BA Physics and BSEngineering, Swarthmore College, PA, USA;MPhil Engineering, U of Cambridge, UK;PhD Electrical Engineering, MIT, Cambridge,MA, USA. Research: I pursue thedevelopment and application of phase-stableultrafast lasers and related technology.Present projects include time and frequencydistribution with 1 part in 1015 precision andcoherent control with phase-stablefemtosecond lasers.www.physics.ubc.ca/php/directory/research/fac-1p.phtml?entnum=278

New Masterminds: Brain Gains at Science

Avilés Barth Carenini Cytrynbaum Eberhardt Ekeland Graham

SYNERGY » 15

Kevin Leyton-Brown, Assist. Prof., Dept. ofComputer Science; BSc Computer Science,McMaster University, Hamilton, ON,Canada; MSc and PhD Computer Science,Stanford University, CA, USA. Research: Iwork at the intersection of computer scienceand microeconomics, studying computationalproblems in economic contexts and incentiveissues in multi-agent systems. Researchcommunities in which I participate includeartificial intelligence, game theory, empiricalalgorithmics, and electronic commerce.www.cs.ubc.ca/~kevinlb

Evgeny A. Pakhomov, Assist. Prof., Dept. ofEarth & Ocean Sciences; Adj. Prof., U ofFort Hare, South Africa; MSc Hydrobiology,Kazan State University, Russia; PhDBiological Oceanography, P. P. ShirshovInstitute of Oceanology, Russian Academy ofSciences, Moscow. Research: I am abiological oceanographer covering topicsfrom species ecology to ecosystem structureas well as physical–biological andbiochemical coupling. My current researchfocusses on changes in the Antarcticecosystem functioning due to global climatechange.www.eos.ubc.ca/public/people/faculty/E.Pakhomov.html

Paul W. Rasmussen, Sr. Instructor, Dept. ofChemistry; BSc Chemistry, AcadiaUniversity, Wolfville, NS, Canada; PhDOrganic Chemistry, McMaster University,Hamilton, ON, Canada. Research: I aminterested in practical applications ofinstrumental methods of analysis, automateddata acquisition and analysis, studentassessment in the undergraduate laboratory,and spectroscopic data storage and retrieval.

Andrew J. Ridgwell, Assist. Prof., Dept. ofEarth & Ocean Sciences; CRC; BA NaturalSciences, U of Cambridge, UK; MScEnvironmental Sciences, U of Nottingham,UK; PhD Earth System Modelling, U of EastAnglia, Norwich, UK. Research: Myresearch involves the development and useof computer models in understanding howthe concentration of carbon dioxide in theatmosphere is controlled, its relationship topast and future changes in climate, and therole played by the evolution of life on Earth.www.eos.ubc.ca/public/people/faculty/A.Ridgwell.html

Matias O. Salibian-Barrera, Assist. Prof.,Dept. of Statistics; BSc Mathematics, U ofBuenos Aires, Argentina; PhD Statistics,UBC, Vancouver, Canada. Research: Idevelop statistical methods that can extractmeaningful information from large data setswhen the quality of the data is doubtful. Ialso work at the intersection of statistics andcomputer science.

Dominik M. Schötzau, Assist. Prof., Dept. ofMathematics; CRC; BSc, MSc and PhDMathematics, ETH Zürich, Switzerland.Research: My main research interest is thedesign of computer methods for thesimulation of complex phenomena in fluidmechanics and electromagnetics. Currentprojects include the development of newfinite element methods for such simulations.www.math.ubc.ca/~schoetzau

Charles J. Thompson, Prof. and Head,Dept. of Microbiology & Immunology; BScWashington and Lee University, Lexington,Virginia, USA; MSc and PhD PennsylvaniaState University, USA. Research:

We carry out molecular genetic studies ofmycobacteria, pathogenic microbes that causetuberculosis and streptomyces, a related groupof multicellular filamentous procaryotes thatundergo a program of colonial differentiationcoordinated with biosynthesis of antibiotics andpharmaceutically important natural compounds.www.microbiology.ubc.ca

Geoffrey O. Wasteneys, Assoc. Prof., Dept.of Botany; CRC; BSc Biology, CarletonUniversity, Ottawa, ON, Canada; PhD PlantCell Biology, Australian National University,Canberra. Research: Plant shape dependson the cell wall, whose mechanicalproperties are modulated by cytoskeletalprotein networks. My team combinesmolecular genetics and microscopy tounravel the mysteries about how thesenetworks function.www.botany.ubc.ca/wasteneys.htm

Recent appointments also include:Keith Adams, Assist. Prof., Dept. of Botanyand Faculty ofAgricultural Sciences –NeilBalmforth, Prof., Depts. of Mathematics andEarth & Ocean Sciences – Jin-Gui Chen,Assist. Prof., Dept. of Botany – Keng ChangChou, Assist. Prof., Dept. of Chemistry –Marco Ciufolini, Prof., Dept. of Chemistry –Kurt Eiselt, Sr. Instructor, Dept. of ComputerScience – Michael Paul Friedlander, Assist.Prof., Dept. of Computer Science – Darren I.Irwin, Assist. Prof., Dept. of Zoology – RichardKenyon, Prof., Dept. of Mathematics – KevinMurphy, Assist. Prof., Depts. of ComputerScience and Statistics – Rachel Pottinger,Assist. Prof., Dept. of Computer Science –Jeffery Richards, Assist. Prof., Dept. ofZoology – OzgurYilmaz, Assist. Prof., Dept. ofMathematics – Eric Wohlstadter, Assist. Prof.,Dept. of Computer Science – Steve Wolfman,Instructor, Dept. of Computer Science

Pakhomov Rasmussen Ridgwell Salibian-Barrera Schötzau Thompson Wasteneys

(Photos [Ekeland/Ridgwell/Schötzau/Wasteneys]: Bayne Stanley)

16 « SYNERGY

Murray Newman Award, Vancouver Aquarium• David R. Jones, Prof., Zoology

Top 50 Contributors to Science and Technology(2003), Scientific American• Daniel Pauly, Prof., Zoology

Service Award, Geological Association ofCanada• Greg Dipple, Prof., EOS• Catherine Hickson, Adj. Prof., EOS• Steve Rowins, Assist. Prof., EOS

Sproule Annual Achievement Award, CanadianSociety of Unconventional Gas• Marc Bustin, Prof., EOS

W. H. Weston Award, Mycological Society ofAmerica• Mary Berbee, Assoc. Prof., Botany

Lorne WhiteheadUBC’s new VicePresident, Academicand ProvostDr. Whitehead wasappointed VicePresident, Academicand Provost July 1,2004. He joinedthe UBC Dept. ofPhysics & Astro-nomy in 1994, was appointed professor in1999 and has served as associate dean and deanpro tem in the Faculty of Science.

Science Department Heads NewsCharles Thompson assumed the headship of theDept. of Microbiology & Immunology in May,2004. Robert J. Woodham stepped down at theend of 2003 after eight years at the helm of theDept. of Computer Science.

Faculty of Science: Kudos and NewsIn 2003/2004 Science faculty memberswon the following prestigious academicawards.

Anita Borg Early Career Award, ComputingResearch Association• Joanna McGrenere, Assist. Prof., ComputerScience

Killam Research Fellowship, Canada Councilfor the Arts • Tony Sinclair, Prof., Zoology

Steacie Fellowship, NSERC• Patrick Keeling, Assist. Prof., Botany

Fellow of the Mineralogical Society ofAmerica• Lee Groat, Prof., Earth & Ocean Sciences(EOS)

Fellow of the American Physical Society • Mona Berciu, Assist. Prof., Physics &Astronomy

• Ian Affleck, Prof., Physics & Astronomy

Foreign Honorary Fellow of the AmericanAcademy of Arts and Science• Walter Hardy, Prof., Physics & Astronomy

Honorary Doctorate, University of Guelph• Julian Davies, Prof. Em., Microbiology &Immunology

Academic Gold Medal for Students, GovernorGeneral• Mirela Andronescu, Computer Science

Andre-Aisenstadt Prize, Centre de RecherchesMathématiques (CRM)• Nike Vatsal, Prof., Mathematics

Prix de l'IHP, Institut Henri Poincaré• Gordon Slade (co-winner), Prof.,Mathematics

Ribenboim Prize, Canadian Number TheoryAssociation• Michael Bennett, Assoc. Prof.,Mathematics

CAP-CRM Prize in Theoretical Physics,Canadian Association of Physicists & CRM • Matthew Choptuik, Prof., Physics &Astronomy

• Jeremy Heyl, Assist. Prof., Physics &Astronomy

Best Paper Award, Association forComputing Machinery (ACM-SIGMOD)• Raymond Ng & Yuhan Cai, ComputerScience

Coxeter-James Prize, Canadian MathematicalSociety• Izabella Laba, Assoc. Prof., Mathematics

Rollo Davidson Prize, Cambridge• Alexander Holroyd (co-winner), Assist. Prof.,Mathematics

Award for Pure or Applied InorganicChemistry, Chemical Institute of Canada• Michael Wolf, Assoc. Prof., Chemistry

Catalysis Award, Chemical Institute of Canada• Colin Fyfe, Prof., Chemistry

Distinguished Academics of the Year Award,Confederation of University FacultyAssociations of BC (CUFA)• David H. Dolphin, Prof., Chemistry

Glenn T. Seaborg Award, American ChemicalSociety (ACS-NCD)• Donald G. Fleming, Prof., Chemistry

Lecture Award, Merck Frosst Centre forTherapeutic Research • Martin Tanner, Prof., Chemistry

Distinguished Achievement Award, AmericanStatistical Association (ASA-ENVR), andHunter Lecture, International EnvironmetricsSociety• James V. Zidek, Prof., Statistics

Dobzhansky Prize, International Society forthe Study of Evolution, and Howard AlperPostdoctoral Prize, NSERC• Aniel Agrawal, PDF Zoology

Return undeliverable Canadianaddresses to:Faculty of Science (SYNERGY)1505–6270 University BoulevardVancouver BC V6T 1Z4. Canada

SYNERGY » JOURNAL OF UBC SCIENCEPublished by the Faculty of ScienceUniversity of British Columbia (UBC)www.science.ubc.ca Fax. 604.822.5558E-mail. synergy.science@ubc.caEditor. Carola Hibsch-Jetter, UBC Science CommunicationsWriters. Research articles: Mari-Louise Rowley,Pro-Textual CommunicationsCommunity features: Brian Marcus (UBC, Math),Carola Hibsch-JetterCirculation. 30,000 (two issues per year)Layout/Art Work. www.mediagroup.ubc.caPhotos. Provided by Faculty of Science or as indicated. All rights reserved.

Printed in Canada.

40589566