Bringing chemical data onto the Semantic Web.

Present chemical data storage methodologies place many restrictions on the use of the stored data. The absence of sufficient high-quality metadata prevents intelligent computer access to the data without human intervention. This creates barriers to the automation of data mining in activities such as quantitative structure-activity relationship modelling. The application of Semantic Web technologies to chemical data is shown to reduce these limitations. The use of unique identifiers and relationships (represented as uniform resource identifiers, URIs, and resource description framework, RDF) held in a triplestore provides for greater detail and flexibility in the sharing and storage of molecular structures and properties.

Behaviour of small solutes and large drugs in a lipid bilayer from computer simulations.

To reach their biological target, drugs have to cross cell membranes, and understanding passive membrane permeation is therefore crucial for rational drug design. Molecular dynamics simulations offer a powerful way of studying permeation at the single molecule level. Starting from a computer model proven to be able to reproduce the physical properties of a biological membrane, the behaviour of small solutes and large drugs in a lipid bilayer has been studied. Analysis of dihedral angles shows that a few nano seconds are sufficient for the simulations to converge towards common values for those angles, even if the starting structures belong to different conformations. Results clearly show that, despite their difference in size, small solutes and large drugs tend to lie parallel to the bilayer normal and that, when moving from water solution into biomembranes, permeants lose degrees of freedom. This explains the experimental observation that partitioning and permeation are highly affected by entropic effects and are size-dependent. Tilted orientations, however, occur when they make possible the formation of hydrogen bonds. This helps to understand the reason why hydrogen bonding possibilities are an important parameter in cruder approaches which predict drug absorption after administration. Interestingly, hydration is found to occur even in the membrane core, which is usually considered an almost hydrophobic region. Simulations suggest the possibility for highly polar compounds like acetic acid to cross biological membranes while hydrated. These simulations prove useful for drug design in rationalising experimental observations and predicting solute behaviour in biomembranes.

Prediction of properties from simulations: a re-examination with modern statistical methods.

We discuss models fit to data collected by Duffy and Jorgensen to predict solvation free energies and partition equilibria of drugs, organic molecules, aromatic heterocycles, and other molecules. These data were originally examined using linear regression, but here more recently developed statistical models are applied. The data set is complicated due to the presence of discrepant observations and also curvature in the response. In some cases it is possible to discard a small number of the observations to get good fit to the data, but, in others, discarding an increasing proportion of the observations does not improve the fit. Our general preference is to use robust parameter estimation which downweights to reduce the influence of discrepant observations on the fitted models. Models are selected for four responses using linear or more complicated representations of the explanatory variables, such as cubic polynomials, B-splines, or smoothers via generalized additive models (GAMs). Variables are chosen using the traditional approach of formal tests to assess contribution to the fit of a model, and resampling methods including bootstrap are also considered to assess the prediction error for given models. Results of our analysis indicate that GAMs are an improvement on linear models for describing the data and making predictions. In general robust regression models and GAMs have the smallest conditional expected loss of prediction over the four responses. In addition, robust regression models offer the advantage of identifying molecules that perform poorly in the fit. In general, models were identified that yielded an improvement of approximately 50% in the conditional expected loss of prediction compared with the original parametrization of Duffy and Jorgensen. It was also found that the use of cross-validation to compare models was unreliable, and bootstrapping is preferred.

Computer simulation of small molecule permeation across a lipid bilayer: dependence on bilayer properties and solute volume, size, and cross-sectional area.

Cell membrane permeation is required for most drugs to reach their biological target, and understanding this process is therefore crucial for rational drug design. Recent molecular dynamics simulations have studied the permeation of eight small molecules through a phospholipid bilayer. Unlike experiments, atomistic simulations allow the direct calculation of diffusion and partition coefficients of solutes at different depths inside a lipid membrane. Further analyses of the simulations suggest that solute diffusion is less size-dependent and solute partitioning more size-dependent than was commonly thought.

A review of protein-small molecule docking methods.

The binding of small molecule ligands to large protein targets is central to numerous biological processes. The accurate prediction of the binding modes between the ligand and protein, (the docking problem) is of fundamental importance in modern structure-based drug design. An overview of current docking techniques is presented with a description of applications including single docking experiments and the virtual screening of databases.

The configurational dependence of binding free energies: a Poisson-Boltzmann study of Neuraminidase inhibitors.

The linear finite difference Poisson-Boltzmann (FDPB) equation is applied to the calculation of the electrostatic binding free energies of a group of inhibitors to the Neuraminidase enzyme. An ensemble of enzyme-inhibitor complex conformations was generated using Monte Carlo simulations and the electrostatic binding free energies of subtly different configurations of the enzyme-inhibitor complexes were calculated. It was seen that the binding free energies calculated using FDPB depend strongly on the configuration of the complex taken from the ensemble. This configurational dependence was investigated in detail in the electrostatic hydration free energies of the inhibitors. Differences in hydration energies of up to 7 kcal mol(-1) were obtained for root mean square (RMS) structural deviations of only 0.5 A. To verify the result, the grid size and parameter dependence of the calculated hydration free energies were systematically investigated. This showed that the absolute hydration free energies calculated using the FDPB equation were very sensitive to the values of key parameters, but that the configurational dependence of the free energies was independent of the parameters chosen. Thus just as molecular mechanics energies are very sensitive to configuration, and single-structure values are not typically used to score binding free energies, single FDPB energies should be treated with the same caution.

Biomimetic synthesis of lantibiotics.

The lantibiotics are a class of highly posttranslationally modified small peptide antibiotics containing numerous lanthionine and dehydroamino acid residues. We have prepared peptides containing multiple dehydroamino acids and cysteine residues in order to probe the biomimetic synthesis of the lantibiotics from their precursor peptides. A novel synthetic methodology was developed to allow the synthesis of multiple dehydroamino acid containing peptides. Cyclisations were rapid, quantitative and regiospecific. Remarkably the peptide sequences alone appear to contain sufficient information to direct a series of stereo- and regiospecific ring closures. Thus both the two linear peptides for the B and E-rings closed stereoselectively. In the case of the A-ring precursor peptide which contained two dehydroamino acids, cyclisation was again totally regioselective, although not totally stereoselective.

Molecular dynamics of mouse and Syrian hamster PrP: implications for activity.

Molecular dynamics computer simulations have been performed on Mouse (Mo) and Syrian Hamster (SHa) prion proteins. These proteins differ, primarily, in that the SHa form incorporates additional residues at the C-terminus and also includes a segment of the unstructured N-terminal region that is required for infectivity. The 1-ns simulations have been analyzed by using a combination of dynamical cross-correlation maps, residue-residue contact plots, digital filtering, and residue-based root-mean-square deviations. The results show that the extra residues present in the SHa form at the C- and N-termini produce changes in the stability of key regions of the protein. The loop region between strand S2 and helix B that contains part of the proposed discontinuous binding site for the chaperone, protein X, is found to be more stable in SHa than in the Mo protein; these results are consistent with the NMR data of James et al. (James et al. Proc Natl Acad Sci USA 1997;94:10086-10091). In addition, a degree of flexibility within the region between and including strand S1 and helix A is also shown in SHa, which is not present in the Mo form; the cross-correlation maps suggest that this is a consequence of the additional unstructured N-terminal region. Furthermore, the extra residues in the N-terminal region of SHa are found to form a beta-bridge with the beta-sheet, within which critical point mutations associated with prion diseases lie. The implications of these results for the conformational interconversion pathway of the prion protein are discussed.

Binding constants of neuraminidase inhibitors: An investigation of the linear interaction energy method.

The linear interaction energy (LIE) method has been applied to the calculation of the binding free energies of 15 inhibitors of the enzyme neuraminidase. This is a particularly challenging system for this methodology since the protein conformation and the number of tightly bound water molecules in the active site are known to change for different inhibitors. It is not clear that the basic LIE method will calculate the contributions to the binding free energies arising from these effects correctly. Application of the basic LIE equation yielded an rms error with respect to experiment of 1.51 kcal mol(-1) for the free energies of binding. To determine whether it is appropriate to include extra terms in the LIE equation, a detailed statistical analysis was undertaken. Multiple linear regression (MLR) is often used to determine the significance of terms in a fitting equation; this method is inappropriate for the current system owing to the highly correlated nature of the descriptor variables. Use of MLR in other applications of the LIE equation is therefore not recommended without a correlation analysis being performed first. Here factor analysis was used to determine the number of useful dimensions contained within the data and, hence, the maximum number of variables to be considered when specifying a model or equation. Biased fitting methods using orthogonalized components were then used to generate the most predictive model. The final model gave a q(2) of 0.74 and contained van der Waals and electrostatic energy terms. This result was obtained without recourse to prior knowledge and was based solely on the information content of the data.

Coherent motion detection and letter position encoding.

We identified 24 'good' and 24 'poor' coherent motion detectors from an unselected sample of young adults. The two groups were matched for reading ability, age and IQ. All subjects carried out two tasks in which optimal performance depended on accurate letter position encoding: a lexical decision task and a primed reaction time task. We found that accurate letter position encoding was predicted by performance in the motion detection task. Since coherent motion detection depends on input from the magnocellular pathway, these findings suggest that information carried by the magnocellular system may be required for encoding letter position. Furthermore, these results may have implications for reading disability which is said to be associated with magnocellular dysfunction.

Molecular dynamics simulation of a hydrated phospholipid bilayer.

A hydrated bilayer of the phospholipid 1,2-dimyristoyl-sn-glycero-3-phosphorylcholine (DMPC) has been studied in the course of a molecular dynamics simulation. Comparison of the simulation results with experiment indicates that generally the two agree well. Data are presented concerning all the major system regions, including the hydrocarbon chains, the glycerol region, the lipid headgroups and the hydrating water molecules. The simulations suggest that this model can be extended to the study of more complex systems of greater biochemical interest, such as membrane bound proteins.

Stratospheric aerosol measurements with implications for global climate.

In this paper we present measurement results obtained in New Mexico with bistatic optical probing of the atmosphere using a searchlight beam. The data yield vertical profiles of the aerosol attenuation coefficient. Because they approximate proportionality to aerosol concentration, these profiles provide information concerning the aerosol layer structure and its parameters. During a 9-day period in October and November 1970, a series of forty-one such profiles was obtained which includes altitudes 12-25 km, selected for study because of the relatively high aerosol content of this stratospheric region and its relation to global climate. The mean stratospheric aerosol distribution for this period is double layered with maxima at 15.6 km and 19.3 km. An early phase of volcanic dust incursion is examined. A chronology of stratospheric aerosol concentration levels is developed based on measurements with the same instrumentation at the same sites since February 1963. The chronology shows that the particulate background level of a nonvolcanic, nonpolluted stratosphere is represented by an aerosol optical thickness Tau(p)(O.55 micro) = 2.0 x 10(-2). A calculation with this background level is included, which indicates feasibility of a laser-satellite method for acquiring data related to global climate. The Appendix introduces an original method of numerical integration (used in the calculations), which is suited to radiative transfer studies.