Cancer iis ia idestructive idisease, ithe ifrequency iof iwhich iis igradually iincreasing iall iover ithe iworld. iAlthough iit idiffers idepending ito ithe ikind iof icancer iand iwhere iit ioccurs, icancer imay ibe itreated iif iit idetected iat ithe iright itime. iSo, ithe icorrect itreatment iand ithe iright itechnology ican iprovide icritical ilife isupport ito icancer ipatients. iDetecting i i ithe istage iof iprecancer iis ibased ion isearching ifor imutations iin igene ior iprotein isequences ithat icause icancers.
iGenetic itesting imay ihelp ito ipredict irisks iof ia iparticular itypes iof icancers.
i iGenetic itesting/diagnosis iis iavailable ifor ibreast, iovarian, icolon, ithyroid iand isome iother icancers ibut itill inow ithere iis inot iany itechnique ito idetect iall ipossible itypes iof icancers ithat ioccurred ivia ia ispecific imutation iin iTP53 igene. iThe idefect iin ifunction iof ithe iprotein ip53 ithat iproduced iby iTP53 igene iis ione iof ithe icommon igenetic ialterations iin ihuman icancer, iand iclose ito ihalf iof iall ihuman itumors icarry ip53 igene imutations iwithin itheir icells.
iSo, ithe idiagnosis iof imutations iin igene iis ibased ion ithe iwhole idatabase i(big idataset ienough), ilike iUniversal iMutation iDatabase i(UMDCell-line2010), ito ireach icorrect iresults.
iThe iproblem iis, ithere iare iseveral iprimitive idatabases i(e.g. iExcel igenome iand iprotein idatabase) icontain idataset iof iTP53 igene iwith iits imutations ithat icause idiseases i(cancers), iBut ithis ibig idatabase icannot idetect iand idiagnose icancers, ibecause iit idoes inot ihave ian iefficient idata imining imethod, iwhich ican idiagnose ipre-cancer’s istage. i iHence, ithe igoal iof ithis iresearch iis ito ireach ia idata imining itechnique, ithat iemploys ineural inetwork, iwhich idepends ion ithe ibig idatasets.
iThis iproposed itechnique ihas ibeen idone iby itwo istages, ifirst: ibioinformatics itechniques iby iusing i i iBLAST, iCLUSTALW, iNCBI, iand iFASTA.
iTo iknow iif ithere iare imalignant imutations ior inot. iSecond: idata imining itechnique ito iclassify ithe imalignant imutations iin ipre-cancer iclassification icase, iwhere ithree isub- iFeed iforward iback ipropagation ineural inetworks, ione ifor ieach isub-dataset ihave ibeen iused, iand i(7) iout iof i(53) ifields ifrom idatabase ifields ihave ibeen iselected. iFurthermore, iit igives itraining irate i(1) iand iMean iSquare iError i(0.1E10-10). i iFinally, ithis itechnique iprovides ia iquick, iaccurate, ieffective, ifast, iflexible iand ieasy iway ito idiagnosis ithe ipre- icancers