Structures(from(scratch(Rhiju(Das,(Departments(of(Biochemistry(&(Physics(BIOC(218(Feb(2010(PredicCng(protein(structure!GTPDIIVNAQINSEDENVLDFIIEDEYYLKKRGVGAHIIKVASSPQLRLLYKNAYSTVSCGNYGVLCNLVQNGEYDLNAIMFNCAEIKLNKGQMLFQTKIWR(?(This(will(happen(to(you(a(lot.(Proteins!Proteins!Proteins!Two(fundamental(problems!GTPDIIVNAQINSEDENVLDFIIEDEYYLKKRGVGAHIIKVASSPQLRLLYKNAYSTVSCGNYGVLCNLVQNGEYDLNAIMFNCAEIKLNKGQMLFQTKIWR(?(1.(PredicCng(protein(struc ture((ugcuccuaguacgagaggaccggagug(?(2.(PredicCng(RNA(structure(Driving!innova*on!in!protein!structu re!predic*on:!“CASP”!Critical Assessment of Structure Prediction Five blind predictions per target CASP1((1994)(From!Neil!Clarke,!CASP7!assessor’s!talk!RMSD:!16.0!Å!Driving!innova*on!in!protein!structu re!predic*on:!“CASP”!Critical Assessment of Structure Prediction Five blind predictions per target CASP3((1998)(RMSD:!!4!to!6!Å!DAVID!BAKER!&!colleagues!CASP6!(2004)!T0281 (1.6 Å over 70 residues)!DAVID!BAKER!&!colleagues!De#novo!Modeling!with!RoseTa!Stage I. Fragment Assembly!De#novo!Modeling!with!RoseTa!Stage II. All-atom refinement!Ingredients!of!a!high!resol u*on !po ten*al!1.(Van(der(waals(packing(2.(Hydrogen(bonds(D!3.(ManifestaCons(of(water(The(hydrophobic(effect(The(cost(of(desolvaCon(Polar!atoms!NonVpolar!atoms!4.(Torsional(potenCal!Ingredients!of!a!high!resol u*on !po ten*al!1.(Van(der(waals(packing(2.(Hydrogen(bonds(D!–!Michael!LeviT,!1969!3.(ManifestaCons(of(water(The(hydrophobic(effect(The(cost(of(desolvaCon(Polar!atoms!NonVpolar!atoms!RoseTa!in!ac*on!Na*ve!(CheY)!A!~1000Vfold!increase!in!computa*onal!power!Cα(RMSD(to(naCve(structure(All^atom(energy(A!~1000Vfold!increase!in!computa*onal!power!Na*ve!(CheY)!Lowest!energy!RoseTa!structure!Cα(RMSD(to(naCve(structure(All^atom(energy(RoseTa@home!RoseTa@home!in!CASP7!From!Neil!Clarke,!CASP7!assessor’s!talk!on!“free!modeling”!Number(of(top^3(votes(Expected(by(c hanc e (CASP7(predictors(Expected(by(c hanc e (Number(of(groups(De#novo!successes:!allVβ#Native! Model!2.0 Å over 61 residues CASP7 target T0316 (domain 3)De#novo!successes:!allVα#1.4(Å(over(90(residues(Native! Model!CASP7 target T0283 (112 residues)!De#novo!modeling:!connecCons(to(the(real(world!The!crystallographic!phase!problem!Engineering!new!protein!folds!and!new!enzymes(NonVbiological!polymers:!beta!proteins!(Hype)(Reality!Reality!Native!Is!protein!folding!solved?!Model!Native!NO!((•!Success!in!<1/3!of!cases.!•!Conforma*onal!sampling!s*ll!a!huge!issue!Can!you!pick!out!the!right!one?!T304!(CASP7)!Crystallographic!model!Best!CASP!model!T304!(CASP7)!Can!you!pick!out!the!right!one?!Crystallographic!model!Best!CASP!model!T304!(CASP7)!Can!you!pick!out!the!right!one?!A(symptom(of(poor(conformaConal(sampling(Two(fundamental(problems!ugcuccuaguacgagaggaccggagug(GTPDIIVNAQINSEDENVLDFIIEDEYYLKKRGVGAHIIKVASSPQLRLLYKNAYSTVSCGNYGVLCNLVQNGEYDLNAIMFNCAEIKLNKGQMLFQTKIWR(?(?(1.(PredicCng(protein(structure((2.(PredicCng(RNA(structure(Proteins! RNA!How!a!physicist!got!into!biochemistry!(2000)!A!flourishing!RNA!world!Engineered(ribozymes(and(aptamers(Conserved(non^coding(RNA((Conserved “cloverleaf” RNAs; Human Accelerated Region 1 RNA.Haussler et al. 2006Breaker & colleagues, 2007“Riboswitches”(The(Das(Lab(Goal:(Nucleic(Acid(Structures(You(Can(Trust(ugcuccuaguacgagaggaccggagug!?(With!de#novo!protein!structure!modeling!as!an!inspira*on,!how!far!can!we!get!with!computers?!Words!an d!grammar!for!RNA?!GACACUAAGUUCGGCAUCAAUAUGGUGACCUCCCGGGAGCGGGGGACCACCAGGUUGCCU AGAGGGGUGAACCGGCCCAGGUCGGAAACGGAGCAGGUCAAAACUCCCGUGCUGAUCAGUAGUGU!Signal(RecogniCon(ParCc le(RNA(Oubridge!et!al.,!2002!Words!an d!grammar!for!RNA?!Canonical(double(helices(Non^canonical(regions(Words!an d!grammar!for!RNA?!De#novo!modeling#Fragment!Assembly!of!RNA!(FARNA)!De#novo!modeling#Ingredients!of!a!high!resol u*on !po ten*al!1.(Van(der(waals(packing(2.(Hydrogen(bonds(D!–!Michael!LeviT!“Detailed(molecular(model(of(transfer(RNA”,(Nature#1969.(3.(ManifestaCons(of(water(The(hydrophobic(effect(The(cost(of(desolvaCon(Polar!atoms!NonVpolar!atoms!Does(it(work?(Na*veVstate!discrimina*on!Low!resolu*on!(FARNA)!energy!NaCve^like(conformaCons(Non^naCve(“decoys”(The(most(conserved(region(of(the(signal(recogniC on(par C c le (Na*veVstate!discrimina*on!Low!resolu*on!(FARNA)!energy!NaCve^like(conformaCons(Non^naCve(“decoys”(The(most(conserved(region(of(the(signal(recogniC on(par C c le (High!resolu*on!energy!Na*veVstate!discrimina*on!Low!resolu*on!(FARNA)!energy!NaCve^like(conformaCons(Non^naCve(“decoys”(The(most(conserved(region(of(the(signal(recogniC on(par C c le (High!resolu*on!energy!Can!we!decipher!all!th e!known!“words”?!De#novo!modeling!Na*ve!!!!!!!!!!!!!!!!Model!1.4!Å!rmsd!1.4!Å!rmsd!1.7!Å!rmsd!In!half!the!cases,#de!novo!modeling!achieves!<!2.0!Å!structures,!and!selects!them.!De!novo!modeling!Na*ve!!!!!!!!!!!!!!!!Model!1.4!Å!rmsd!1.4!Å!rmsd!1.7!Å!rmsd!In!half!the!cases,#de!novo!modeling!achieves!<!2.0!Å!structures,!and!selects!them.!De#novo!modeling!The(biggest(bogleneck:(conformaConal(sampling(De#novo!modeling!The(biggest(bogleneck:(conformaConal(sampling(1.0!Å!rmsd!We(know(the(rules(of(the(game((but!we!have! to!play!it!be0er)!A!universal!obsession!BeaC ng(the(“astronomical”(conformaConal(sampling(problem(SoluCon(1?(Data(13!22!32!42!52!62!72!83!13! 22! 32! 42! 52! 62! 72! 83!SHAPE(w/(adenine(3’!5’!13!22!32!42!52!62!72!83!Model( NaCve(SoluCon(2?(Humans(FOLD .IT(Baker!lab!With!UW!Comp.!Sci.!(Adri en !Treuille,!Seth!Cooper,!Zoran!Popovic,!David!Salesin,!others…)!ETERNA(With!Adrien!Treuille!(now!at!CarnegieVMellon)!and!Jeehyung!Lee!SoluCon(3?(Physics(Computa*onally!expensive,!but!gerng!faster!S*ll!no!case!of!blind!predic*ons!of!structure!P P C10!G9!C4!G3!SoluCon(4.(Beger(algorithms(G C A AP P C10!G9!C4!G3!StepVbyVstep!sampli ng!G C A AP P C10!G9!C4!G3!StepVbyVstep!sampli ng!GP P C10!G9!G5!C4!G3!P StepVbyVstep!sampli ng!P P P C10!G9!G5!C4!G3!StepVbyVstep!sampli ng!P P P C10!G9!G5!C4!G4!StepVbyVstep!sampli ng!P P P P C10!G9!A8!G5!C4!G3!StepVbyVstep!sampli ng!P P P P P C10!G9!A8!A7!G5!C4!G3!StepVbyVstep!sampli ng!P P P P P P P C10!G9!A8!A7!C6!G5!C4!G3!StepVbyVstep!sampli ng!1ZIH!NMR!Lowest!Energy!StepVbyVstep!sampli ng!Aha!–!terms!for:!!!•!base!stacking!!!•!RNA!torsional!poten*al!Had!been!dialed!down!to!zero.!(A#legacy#of#fragment#assembly)!1ZIH!NMR!Lowest!Energy!StepVbyVstep!sampli
View Full Document
Unlocking...