Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Standard view
Full view
of .
Look up keyword
Like this
0 of .
Results for:
No results containing your search query
P. 1


Ratings: (0)|Views: 13|Likes:
Published by Nathan McCorkle

More info:

Published by: Nathan McCorkle on Apr 20, 2011
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less





A Biomolecular Computing Method Based onRho Family GTPases
Jian-Qin Liu* and Katsunori Shimohara
 , Member, IEEE 
In this paper, we propose a new biomolecularcomputing method based on Rho family GTPases, and discuss theschemes of representation and operations of molecular computingby Rho family GTPases applied to solve large-scale 3-SAT prob-lems. We also present the optimal condition for the regulationschemes dependent on the temperature, kinase activity, and typesof cells. This work is important for potential implementationof biomolecular computers using Rho family GTPases in whichan optimized controlling scheme can make the best use of theinteractions of signaling pathways in a computing system madeby the large-scale abundance of kinases and phosphatases in cells.
 Index Terms—
Biomolecular computing, signaling pathways of cells, 3-SAT problem solving.
I. I
T THE nanoscale of biomolecular computers, we areexploring an unconventional computing mechanismmade of signaling pathways, for which Rho family GTPasesare biologically plausible objects. Reducing the computing cost(i.e., the molecular number and time) of molecular computersbuilt using DNA, RNA, protein, and other biomolecules isan important task. We propose a new molecular computingmethod based on Rho family GTPases, which differs fromthe Adleman–Lipton paradigm of DNA computing and sur-face-based techniques [1]–[4]. The final target of our research on this method is to employ the signaling pathways of Rhofamily GTPases from
in situ
cells (under certain conditions)in order to build a biomolecular computer with optimizedregulation schemes. The interaction mechanisms of pathwaysin biomolecular computing that we designed and controlledcan systematically guarantee scalability, robustness, and pro-grammability for massive parallel computing processes. Wemainly consider the following two aspects:
 A. Manipulation for Algorithms
Based on analyses of manipulations of biomolecular com-puters and feasibility of implementing techniques in wetware,breakthroughs are imperative for computational moleware.From this viewpoint, emphasis should be placed on working
Manuscript receivedSeptember25, 2002;revisedMarch14, 2003.This worwas supported by the Telecommunications Advancement Organizationof Japanas part of “Research on Human Communication.” This paper was presentedin part at IEEE Nanotechnology Conference, Washington, DC, August 2002.
 Asterisk indicates corresponding author.
*J.-Q. Liu is with ATR Human Information Science Laboratories,2-2-2, Hikaridai, Keihanna Science City, Kyoto 619-0288, Japan (e-mail: jqliu@atr.co.jp).K. Shimohara is with ATR Human Information Science Laboratories, Kyoto,Japan.Digital Object Identifier 10.1109/TNB.2003.813935
out solutions to the potential implementation of molecularcomputers.
 B. Toward Potentially Feasible Solutions of Computing
Applying a molecular computing algorithm to the 3-SATproblem is a good starting point from the viewpoint of unconventional paradigms of biologically (nature)-inspiredinformation processing, in order to explore the parallelismmechanism of nanoscience.II. M
 A. Materials
The materials that we expect to be used in potential im-plementation are kinases in the set of Rho family GTPases[5], their corresponding phosphatases, and related signalingpathways in cells. Using these materials, it is possible to builda biomolecular computer for the general purpose of compu-tation. Without loss of generality, we have studied reversibleprocesses of phosphorylation-dephosphorylation controlled bycorresponding kinases and phosphatases that show switchlikebehavior in logic. The phosphorylation and dephosphorylationstates of these switches represent the true or false values inlogic, or one and zero in binary, respectively. In terms of thesesignaling switches of binary logic, the programs for molecularcomputing can be made from signaling molecules, and theirrunning can be carried out using signal pathways under theregulation of kinases and phosphatases. Therefore, a molecularcomputer can be expected to work out solutions to computationby using cells in the manner described.Molecules for representing the variables in 3-SAT com-putation are constructed in terms of signal transduction. Thetarget molecules of cells for signal transduction under theregulation of Rho family GTPases {Rho, Rac, Cdc42} in theset of {PIP 5-kinase, Rhophilin, Rhotekin, PKN, PRK2, citron,citron-kinase, Rho-kinase, MBS, MLC, p140mDia, p140Sra-1, Por1, PI 3-kinase, S 6-kinase, IQGAP, PAKs, MLK3,MEKK4, MRCKs, WASP, N-WASP, and Ack} are used toencode the combinatorial forms of symbols in the form of structural molecular complexes. From the symbolic functionsof the signaling molecules in the previous set, we can get thealphabet set in the meaning of theoretical computer scienceas,where is PIP -kinase, is Rhophilin, is Rhotekin, is PKN,is PRK2, is citron, is citron-kinase, is Rho-kinase,is MBS, is MLC, is p140mDia, is p140 Sra-1,is Por1, is PI 3-kinase, is S 6-kinase, is IQGAP,is PAKs, is MLK3, is MEKK4, is MRCKs, is WASP,
1536-1241/03$17.00 © 2003 IEEE
Fig. 1. Schematic diagram of regulated pathways for 3-SAT computation.
is N-WASP, and is Ack. These can be used to constructthe different words for the program codes in order to designa kind of machine language for biomolecular computers. Thepreviously mentioned symbols, which are defined as variablesin logic, take the phosphorylation form for the true value andthe dephosphorylation form for the false value. When they aredefined in binary form, the phosphorylation state correspondsto one and dephosphorylation corresponds to zero. The cells areregulated by activating Guanine nucleotide Exchange Factors(GEFs) in the set of {Db1 for Cdc42 and Rho, Lbc for Rho,Lfc for Rho, Lsc for Rho, Tiam for Rac, p115 Rho GEF forRho, Vav for Rac and Cdc42 and Rho, Fgdl for Cdc42, Trio forRac and Rho, Ost for Rho and cdc42, Bcr for Rac and Cdc42and Rho, Pix for Rac, and Smg GDS for Rac and Cdc42 andRho} according to the biological evidence reported in [5]. It isbiologically faithful to use these signaling molecules to set thetarget molecules as the states that fit the available operations onpathways in cells. The intracell communication mechanism of related signaling molecules encoded for molecular computingis set into the active state, guided by related pathways. If allof the pathways activated for computation have reached theirstable stages, the interactions of pathways produce the finalresult of computation from partial solutions in underlyingsubspace rather than the whole searching space. Therefore,reasonable efficiency can be achieved from the cut in controlcost. The pathways operate with interactions among them in away functionally equivalent to recursive computation in theory.The whole computing process by biological pathways is car-ried out by phosphorylation–dephosphorylation encoding forthe objects, kinases–phosphatase pathways for candidate pro-duction, and regulation of the interactions of the signaling path-ways of Rho family GTPases in cells. The pathways correspondto the clauses in 3-SAT computation. The running of pathwaysis carried out until the final outputs are determined, wherein themolecular complexes produced by the pathways can satisfy allconditions of the 3-SAT problem. After the pathway productionprocesses, the molecular complexes that satisfy those clauseswill be output for readout. The entire computing process of thebiological pathways is given in Fig. 1.Immunofluorescence analysis is potentially a direct way toreadout the result of molecular computing based on Rho familyGTPases, since it is an available technique in biological bench-works and is practical as well, according to [5]
 B. Operations
The high-level operations for 3-SAT are composed of theoperation set of , where refers tothe operation that activates the pathways for clause constraintfor the 3-SAT computation. refers to the running opathways, and this consists of biochemical reactions guided bydesigned kinases–phosphatases schemes of pathway control,where a labeling process on object molecules and relateddetection is included as well. refers to readout for thefinal result based on detection of the solution from differentpathways by the threshold for common outputs.The major biological molecular operations include the fol-lowing:1) phosphorylation;2) dephosphorylation;3) regulation of kinases for phosphorylation;4) regulation of phosphatases for dephosphorylation;5) activation of the kinase-guide pathways;6) activation of the phosphatase-guided pathways;7) activation of the designed pathways that are guided bykinases, phosphatases, and other related molecular com-plexes and chemicals where labeling molecules are in-volved;8) concentration detection and sieving by a certainthreshold;9) synchronizationbasedonproductsofdifferentpathwaysfor successive pathway reactions (for potential condi-tions of cells
in situ
);10) readout by immunofluorescence analysis.

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->