?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.relation=http%3A%2F%2Fmiis.maths.ox.ac.uk%2Fmiis%2F748%2F&rft.title=Classification+of+Two-Dimensional+Gas+Chromatography+Data&rft.creator=Croci%2C+Matteo&rft.creator=Morawiecki%2C+P.&rft.creator=Prater%2C+John&rft.creator=Sulzer%2C+Valentin&rft.creator=Theil%2C+Florian&rft.subject=Aerospace+and+defence&rft.description=Gas+chromatography+(GC)+is+a+popular+tool+for+chemical+analysis.+Some+samples+are+so+complex+that+a+single+column+does+not+have+enough+power+to+separate+all+of+the+analytes.+In+this+instance+a+higher+resolution+GC+method%2C+known+as+comprehensive+two-dimensional+gas+chromatography+(GCxGC)%2C+is+used.+DSTL+want+to+be+able+to+use+data+from+GCxGC+to+attribute+samples+to+a+particular+region+or+cultivar.+However%2C+the+nature+of+the+data+means+that+several+difficulties+must+be+overcome+before+being+able+to+do+this%3A+noise+from+sample%2C+peak+mis-alignment%2C+and+low+quantity+of+samples.+In+this+report%2C+we+investigate+several+methods+to+overcome+such+difficulties%2C+and+then+classify+the+data.+We+are+very+successful+in+telling+apart+blanks+from+seeds%2C+but+obtain+limited+success+when+trying+to+classify+between+seeds.+The+method+that+shows+the+most+promise+is+k-Nearest+Neighbours+classification+by+Wasserstein+distance.+However%2C+this+is+still+quite+sensitive+to+the+noise+created+by+the+solvent+in+the+sample.+Thus%2C+we+suggest+that+more+blank+runs+be+obtained%2C+so+that+the+%E2%80%98ground+truth%E2%80%99+behaviour+of+the+solvent+is+better+understood%2C+allowing+us+to+remove+the+effect+of+the+solvent+from+seed+data.+We+also+hope+that+the+methods+explored+here+will+be+more+successful+on+the+full+raw+data+than+they+were+on+the+limited+%E2%80%98peaks%E2%80%99+data+available+to+us+for+the+purpose+of+this+study.&rft.date=2017&rft.type=Study+Group+Report&rft.type=NonPeerReviewed&rft.format=application%2Fpdf&rft.language=en&rft.identifier=http%3A%2F%2Fmiis.maths.ox.ac.uk%2Fmiis%2F748%2F1%2FDSTL_StudyGroupReport_v1.pdf&rft.identifier=++Croci%2C+Matteo+and+Morawiecki%2C+P.+and+Prater%2C+John+and+Sulzer%2C+Valentin+and+Theil%2C+Florian++(2017)+Classification+of+Two-Dimensional+Gas+Chromatography+Data.++%5BStudy+Group+Report%5D+++++