Computer modelling studies on the mechanism of action of ribonuclease T1

The mechanism of action of ribonuclease (RNase) T1 is still a matter of considerable debate as the results of x-ray, 2-D nmr and site-directed mutagenesis studies disagree regarding the role of the catalytically important residues. Hence computer modelling studies were carried out by energy minimisation of the complexes of RNase T1 and some of its mutants (His40Ala, His40Lys, and Glu58Ala) with the substrate guanyl cytosine (GpC), and of native RNase T1 with the reaction intermediate guanosine 2',3'-cyclic phosphate (G greater than p). The puckering of the guanosine ribose moiety in the minimum energy conformer of the RNase T1-GpC (substrate) complex was found to be O4'-endo and not C3'-endo as in the RNase T1-3'-guanylic acid (inhibitor/product) complex. A possible scheme for the mechanism of action of RNase T1 has been proposed on the basis of the arrangement of the catalytically important amino acid residues His40, Glu58, Arg77, and His92 around the guanosine ribose and the phosphate moiety in the RNase T1-GpC and RNase T1-G greater than p complexes. In this scheme, Glu58 serves as the general base group and His92 as the general acid group in the transphosphorylation step. His40 may be essential for stabilising the negatively charged phosphate moiety in the enzyme-transition state complex.

A homotopy approach for stabilizing single-input systems with control structure constraints

A linear state feedback gain vector used in the control of a single input dynamical system may be constrained because of the way feedback is realized. Some examples of feedback realizations which impose constraints on the gain vector are: static output feedback, constant gain feedback for several operating points of a system, and two-controller feedback. We consider a general class of problems of stabilization of single input dynamical systems with such structural constraints and give a numerical method to solve them. Each of these problems is cast into a problem of solving a system of equalities and inequalities. In this formulation, the coefficients of the quadratic and linear factors of the closed-loop characteristic polynomial are the variables. To solve the system of equalities and inequalities, a continuous realization of the gradient projection method and a barrier method are used under the homotopy framework. Our method is illustrated with an example for each class of control structure constraint.

Computer modelling studies of ribonuclease A-pyrimidine nucleotide complexes

Different modes of binding of pyrimidine monophosphates 2'-UMP, 3'-UMP, 2'-CMP and 3'-CMP to ribonuclease (RNase) A are studied by energy minimization in torsion angle and subsequently in Cartesian coordinate space. The results are analysed in the light of primary binding sites. The hydrogen bonding pattern brings out roles for amino acids such as Asn44 and Ser123 apart from the well known active site residues viz., His12,Lys41,Thr45 and His119. Amino acid segments 43-45 and 119-121 seem to be guiding the ligand binding by forming a pocket. Many of the active site charged residues display considerable movement upon nucleotide binding.

Computer modeling studies on the binding of 2',5'-linked dinucleoside phosphates to ribonuclease T1-influence of subsite interactions on the substrate specificity

The modes of binding of Gp(2',5')A, Gp(2',5')C, Gp(2',5')G and Gp(2',5')U to RNase T1 have been determined by computer modelling studies. All these dinucleoside phosphates assume extended conformations in the active site leading to better interactions with the enzyme. The 5'-terminal guanine of all these ligands is placed in the primary base binding site of the enzyme in an orientation similar to that of 2'-GMP in the RNase T1-2'-GMP complex. The 2'-terminal purines are placed close to the hydrophobic pocket formed by the residues Gly71, Ser72, Pro73 and Gly74 which occur in a loop region. However, the orientation of the 2'-terminal pyrimidines is different from that of 2'-terminal purines. This perhaps explains the higher binding affinity of the 2',5'-linked guanine dinucleoside phosphates with 2'-terminal purines than those with 2'-terminal pyrimidines. A comparison of the binding of the guanine dinucleoside phosphates with 2',5'- and 3',5'-linkages suggests significant differences in the ribose pucker and hydrogen bonding interactions between the catalytic residues and the bound nucleoside phosphate implying that 2',5'-linked dinucleoside phosphates may not be the ideal ligands to probe the role of the catalytic amino acid residues. A change in the amino acid sequence in the surface loop region formed by the residues Gly71 to Gly74 drastically affects the conformation of the base binding subsite, and this may account for the inactivity of the enzyme with altered sequence i.e., with Pro, Gly and Ser at positions 71 to 73 respectively. These results thus suggest that in addition to recognition and catalytic sites, interactions at the loop regions which constitute the subsite for base binding are also crucial in determining the substrate specificity.

Algorithms for Exact Structure Discovery in Bayesian Networks

Bayesian networks are compact, flexible, and interpretable representations of a joint distribution. When the network structure is unknown but there are observational data at hand, one can try to learn the network structure. This is called structure discovery. This thesis contributes to two areas of structure discovery in Bayesian networks: space--time tradeoffs and learning ancestor relations. The fastest exact algorithms for structure discovery in Bayesian networks are based on dynamic programming and use excessive amounts of space. Motivated by the space usage, several schemes for trading space against time are presented. These schemes are presented in a general setting for a class of computational problems called permutation problems; structure discovery in Bayesian networks is seen as a challenging variant of the permutation problems. The main contribution in the area of the space--time tradeoffs is the partial order approach, in which the standard dynamic programming algorithm is extended to run over partial orders. In particular, a certain family of partial orders called parallel bucket orders is considered. A partial order scheme that provably yields an optimal space--time tradeoff within parallel bucket orders is presented. Also practical issues concerning parallel bucket orders are discussed. Learning ancestor relations, that is, directed paths between nodes, is motivated by the need for robust summaries of the network structures when there are unobserved nodes at work. Ancestor relations are nonmodular features and hence learning them is more difficult than modular features. A dynamic programming algorithm is presented for computing posterior probabilities of ancestor relations exactly. Empirical tests suggest that ancestor relations can be learned from observational data almost as accurately as arcs even in the presence of unobserved nodes.

Characterization of substrate UpA binding to RNase A--computer modelling and energetics approach.

In the past two decades RNase A has been the focus of diverse investigations in order to understand the nature of substrate binding and to know the mechanism of enzyme action. Although this system is reasonably well characterized from the view point of some of the binding sites, the details of interactions in the second base binding (B2) site is insufficient. Further, the nature of ligand-protein interaction is elucidated generally by studies on RNase A-substrate analog complexes (mainly with the help of X-ray crystallography). Hence, the details of interactions at atomic level arising due to substrates are inferred indirectly. In the present paper, the dinucleotide substrate UpA is fitted into the active site of RNase A Several possible substrate conformations are investigated and the binding modes have been selected based on Contact Criteria. Thus identified RNase A-UpA complexes are energy minimized in coordinate space and are analysed in terms of conformations, energetics and interactions. The best possible ligand conformations for binding to RNase A are identified by experimentally known interactions and by the energetics. Upon binding of UpA to RNase A the changes associated,with protein back bone, Side chains in general and at the binding sites in particular are described. Further, the detailed interactions between UpA and RNase A are characterized in terms of hydrogen bonds and energetics. An extensive study has helped in interpreting the diverse results obtained from a number of experiments and also in evaluating the extent of changes the protein and the substrate undergo in order to maximize their interactions.

Modes of binding of 2'-AMP to RNase T1. A computer modeling study

The modes of binding of adenosine 2'-monophosphate (2'-AMP) to the enzyme ribonuclease (RNase) T1 were determined by computer modelling studies. The phosphate moiety of 2'-AMP binds at the primary phosphate binding site. However, adenine can occupy two distinct sites--(1) The primary base binding site where the guanine of 2'-GMP binds and (2) The subsite close to the N1 subsite for the base on the 3'-side of guanine in a guanyl dinucleotide. The minimum energy conformers corresponding to the two modes of binding of 2'-AMP to RNase T1 were found to be of nearly the same energy implying that in solution 2'-AMP binds to the enzyme in both modes. The conformation of the inhibitor and the predicted hydrogen bonding scheme for the RNase T1-2'-AMP complex in the second binding mode (S) agrees well with the reported x-ray crystallographic study. The existence of the first mode of binding explains the experimental observations that RNase T1 catalyses the hydrolysis of phosphodiester bonds adjacent to adenosine at high enzyme concentrations. A comparison of the interactions of 2'-AMP and 2'-GMP with RNase T1 reveals that Glu58 and Asn98 at the phosphate binding site and Glu46 at the base binding site preferentially stabilise the enzyme-2'-GMP complex.

The structures of bacteriorhodopsin with different retinal-Schiff base orientations--computer modeling and energy minimization studies

Bacteriorhodopsin has been the subject of intense study in order to understand its photochemical function. The recent atomic model proposed by Henderson and coworkers based on electron cryo-microscopic studies has helped in understanding many of the structural and functional aspects of bacteriorhodopsin. However, the accuracy of the positions of the side chains is not very high since the model is based on low-resolution data. In this study, we have minimized the energy of this structure of bacteriorhodopsin and analyzed various types of interactions such as - intrahelical and interhelical hydrogen bonds and retinal environment. In order to understand the photochemical action, it is necessary to obtain information on the structures adopted at the intermediate states. In this direction, we have generated some intermediate structures taking into account certain experimental data, by computer modeling studies. Various isomers of retinal with 13-cis and/or 15-cis conformations and all possible staggered orientations of Lys-216 side chain were generated. The resultant structures were examined for the distance between Lys-216-schiff base nitrogen and the carboxylate oxygen atoms of Asp-96 - a residue which is known to reprotonate the schiff base at later stages of photocycle. Some of the structures were selected on the basis of suitable retinal orientation and the stability of these structures were tested by energy minimization studies. Further, the minimized structures are analyzed for the hydrogen bond interactions and retinal environment and the results are compared with those of the minimized rest state structure. The importance of functional groups in stabilizing the structure of bacteriorhodopsin and in participating dynamically during the photocycle have been discussed.

Lazy structure sharing for query optimization

We study lazy structure sharing as a tool for optimizing equivalence testing on complex data types, We investigate a number of strategies for implementing lazy structure sharing and provide upper and lower bounds on their performance (how quickly they effect ideal configurations of our data structure). In most cases when the strategies are applied to a restricted case of the problem, the bounds provide nontrivial improvements over the naive linear-time equivalence-testing strategy that employs no optimization. Only one strategy, however, which employs path compression, seems promising for the most general case of the problem.

Tree structure for efficient data mining using rough sets

In data mining, an important goal is to generate an abstraction of the data. Such an abstraction helps in reducing the space and search time requirements of the overall decision making process. Further, it is important that the abstraction is generated from the data with a small number of disk scans. We propose a novel data structure, pattern count tree (PC-tree), that can be built by scanning the database only once. PC-tree is a minimal size complete representation of the data and it can be used to represent dynamic databases with the help of knowledge that is either static or changing. We show that further compactness can be achieved by constructing the PC-tree on segmented patterns. We exploit the flexibility offered by rough sets to realize a rough PC-tree and use it for efficient and effective rough classification. To be consistent with the sizes of the branches of the PC-tree, we use upper and lower approximations of feature sets in a manner different from the conventional rough set theory. We conducted experiments using the proposed classification scheme on a large-scale hand-written digit data set. We use the experimental results to establish the efficacy of the proposed approach. (C) 2002 Elsevier Science B.V. All rights reserved.

Structural Alignment based Kernels for Protein Structure Classification

Structural alignments are the most widely used tools for comparing proteins with low sequence similarity. The main contribution of this paper is to derive various kernels on proteins from structural alignments, which do not use sequence information. Central to the kernels is a novel alignment algorithm which matches substructures of fixed size using spectral graph matching techniques. We derive positive semi-definite kernels which capture the notion of similarity between substructures. Using these as base more sophisticated kernels on protein structures are proposed. To empirically evaluate the kernels we used a 40% sequence non-redundant structures from 15 different SCOP superfamilies. The kernels when used with SVMs show competitive performance with CE, a state of the art structure comparison program.

Correlations between structure and thermodynamic properties of spinels and their applications in technology

Measurements a/the Gibbs' energy enthalpy and entrupy vffarmation oj chromites, vanadites and alumlnat.:s 0/ F", Ni. Co'. Mn, Zn Mg and Cd, using solid oxide galvanic cells over a ternperature range extending approximately lOOO°C, have shown that the '~'Ilir"!,,, J'JrIl/iJ~ tion 0/ cubic 2-3 oxide spinel phases (MX!O,), from component oxide (MO) with rock-salt and X.Os whir c(1f'l/!ldwn st!'llt'lw,·. call b,' represented by a semi-empirical correlalion, ~S~ = --LiS + L'i,SM +~S~:"d(±O.3) cal.deg-1 mol-1 where /',.SM Is the entropy 0/calian mixing oillhe tetrahedral alld octahedral sites o/the spinel and Sr:~ is tlie enfropy associaf,'d Wifh Ih,' randomization a/the lahn-Telier distortions. A review a/the methods/or evaluating the cation distriblltion lfl spille!s suggeJ{j' l/r,l! Ihe most promising scheme is based Oil octahedral site preference energies from the crystal field theory for the Iral1silioll IIIl'f"! IlIIL';. For I/""-Irallsifioll melal cal ions site preference energies are derived relative /0 thol'lt fLI, [ransilion metal ions from measured high tClllP('ftJi ure Cal iUlI disll iiJuriol1 in spine! phases thar contail! one IransilioJl metal and another non-transition metal carion. For 2-3 srinds compulatiorrs b,IS"J Oil i.!c[J;' Temkin mixing on each catioll subialtice predici JistributionJ that are In fair agreement with X-ray and 1I1'IIIrOll ditTraction, /IIdg""!ic dll.! electrical propcrries, and spectroscopic measurements. In 2-4 spineis mixing vI ions do not foliow strictly ideal slllIistli:al Jaws, Th,' OIl/up) associated with the randomizalion 0/the Jllhn-Teller dislOriioll" appear to be significant, only ill spinels witll 3d'. 3d', 3d' (ifld~UI' iOtls in tetrahedral and 3d' and 3d9 ions in octahedral positions. Application 0/this structural model for predicting the thermodynamic proputies ofspinel solid .,olutiofl5 or,' illustrated. F,lr complex systems additional contributions arising from strain fields, redox equilibria and off-center ions have to be qllalllififti. The entropy correlation for spinels provides a method for evaluating structure tran:.jormafiofl entropies in silllple o.\id.-s, ["founlllion on the relative stabilities ofoxides in different crystallCtructures is USe/III for computer ea/culaliof! a/phase dfugrullls ofIlIrer,',,1 III (N.lll1ie5 by method, similar to thost: used by Kaufman and Bernstein for refractory alloy systems. Examples oftechnoiogical appliCation tnclude the predictioll ofdeoxidation equilibria in Fe-Mn-AI-O s),slelll at 1600°C duj ,'Ulllpltfalion 0/phase relutions in Fe-Ni-Cr-S system,

Use of Vertex Index in Structure-Activity Analysis and Design of Molecules

The last few decades have witnessed application of graph theory and topological indices derived from molecular graph in structure-activity analysis. Such applications are based on regression and various multivariate analyses. Most of the topological indices are computed for the whole molecule and used as descriptors for explaining properties/activities of chemical compounds. However, some substructural descriptors in the form of topological distance based vertex indices have been found to be useful in identifying activity related substructures and in predicting pharmacological and toxicological activities of bioactive compounds. Another important aspect of drug discovery e. g. designing novel pharmaceutical candidates could also be done from the distance distribution associated with such vertex indices. In this article, we will review the development and applications of this approach both in activity prediction as well as in designing novel compounds.

Exploiting the Structure of the Constraint Graph for Efficient Points-to Analysis

Points-to analysis is a key compiler analysis. Several memory related optimizations use points-to information to improve their effectiveness. Points-to analysis is performed by building a constraint graph of pointer variables and dynamically updating it to propagate more and more points-to information across its subset edges. So far, the structure of the constraint graph has been only trivially exploited for efficient propagation of information, e.g., in identifying cyclic components or to propagate information in topological order. We perform a careful study of its structure and propose a new inclusion-based flow-insensitive context-sensitive points-to analysis algorithm based on the notion of dominant pointers. We also propose a new kind of pointer-equivalence based on dominant pointers which provides significantly more opportunities for reducing the number of pointers tracked during the analysis. Based on this hitherto unexplored form of pointer-equivalence, we develop a new context-sensitive flow-insensitive points-to analysis algorithm which uses incremental dominator update to efficiently compute points-to information. Using a large suite of programs consisting of SPEC 2000 benchmarks and five large open source programs we show that our points-to analysis is 88% faster than BDD-based Lazy Cycle Detection and 2x faster than Deep Propagation. We argue that our approach of detecting dominator-based pointer-equivalence is a key to improve points-to analysis efficiency.

Comparative Study of Protein Unfolding in Aqueous Urea and Dimethyl Sulfoxide Solutions: Surface Polarity, Solvent Specificity, and Sequence of Secondary Structure Melting

Elucidation of possible pathways between folded (native) and unfolded states of a protein is a challenging task, as the intermediates are often hard to detect. Here, we alter the solvent environment in a controlled manner by choosing two different cosolvents of water, urea, and dimethyl sulfoxide (DMSO) and study unfolding of four different proteins to understand the respective sequence of melting by computer simulation methods. We indeed find interesting differences in the sequence of melting of alpha helices and beta sheets in these two solvents. For example, in 8 M urea solution, beta-sheet parts of a protein are found to unfold preferentially, followed by the unfolding of alpha helices. In contrast, 8 M DMSO solution unfolds alpha helices first, followed by the separation of beta sheets for the majority of proteins. Sequence of unfolding events in four different alpha/beta proteins and also in chicken villin head piece (HP-36) both in urea and DMSO solutions demonstrate that the unfolding pathways are determined jointly by relative exposure of polar and nonpolar residues of a protein and the mode of molecular action of a solvent on that protein.