Abstract

In this paper, a novel and effective method for impulse noise removal in corrupted color images is discussed. The new method consists of two phases. The first phase is a noise detection phase where a modified Hopfield neural network is used to detect impulse noise pixels. The second is a noise filtering phase where the disadvantage of taking Vector Median in a single color space is addressed and a new algorithm based on performing Vector Median first in RGB space and then in HSI space is presented. The results of simulations performed on a set of standard test images on a wide range of noise corruption show that the proposed method is capable of detecting all the impulse noise pixels with almost zero false positive rates and removes noise while retaining finer image details. It outperforms the standard procedures and is yet simple and suitable for real time applications.

Abstract

Protonated base pairs constitute an important class of RNA base pairing systems, several of which nucleate higher order structures and participate in tertiary interactions joining distant regions in RNA structures. These base pairs are involved in the formation of functionally important regions of RNA such as ribozyme cleavage sites, protein recognition sites, and rRNA motifs. Our earlier report [1] on Hartree Fock level quantum chemical studies on the geometries and interaction energies of a few protonated base pairs, in crystal contexts and gas phase optimized geometries, shows that these base pairs have high interaction energies and display stable geometries. Here we report results of higher level ab initio studies on an extended set of protonated base pairs. The structures of 18 systems, with one of the bases being protonated, were optimized at B3LYP/cc-pVTZ level and their interaction energies were calculated at MP2/aug-cc-pVDZ level. To mimic the solvent effect COSMO calculations have also been carried out on some the systems. The interaction patterns of protonated base pairs are highly diverse, with gas-phase optimized interaction energies in the range from -24 to -49 kcal/mol. The interaction energy values, for all base pairs studied, were found to be dominated by the HF term. This suggests that, in contrast with observations on nonprotonated base pairs, interaction energies of protonated base pairs have higher electrostatic contribution with noticeably reduced electron correlation component. As is also indicated by Mulliken population analysis, Morokuma decomposition analysis of interaction energies shows charge transfer as one of the dominant components in these base pair systems.

Abstract

Riboswitches are ligand dependent cis-acting regulatory elements found in diverse species, in some of their mRNA transcripts in the 5’-untranslated regions (UTRs). They are composed of two distinct, yet mutually interacting, domains: a ‘ligand-binding’ or aptamer domain and an expression platform involved in gene regulation. Conformational changes due to ligand binding affect the folding patterns in the expression platform, enabling riboswitches to exercise a variety of ligand dependent controls on downstream gene expressions. Purine riboswitches have been most widely investigated computationally as well as experimentally, for their remarkable ligand (adenine/guanine) specificity and variation in functional mechanism, despite their small size and their sequence similarities. Though the crystal structures of the ligand bound CLOSED states of the aptamer domains of two of the adenine and guanine riboswitches are known, the characterization of the ligand-free OPEN state is limited primarily to results derived from chemical probing, NMR, fluorescence, single labeled FRET and single molecule force spectroscopy experiments. Consequently, a complete molecular level understanding of mechanism of ligand binding, ligand discrimination and switching is still limited. We have addressed this issue for the aptamer domain of the add A-riboswitch by carrying out ab initio quantum chemical studies on the geometries and interaction energies of ligand-bound binding pocket1 and by studying explicit-solvent molecular dynamics simulations of the adenine riboswitch structure (1Y26), both in the adenine-bound (CLOSED) state and in the adenine-free (OPEN) state.2 Dynamic cross correlation (DCC), root mean square fluctuation (RMSF) and backbone torsion angle analyses of the two trajectories and analysis of the average OPEN and CLOSED state structures reveal, alternative hydrogen bonding schemes involving bases that are not directly involved in ligand binding, but, which constitute the lining of the binding pocket and possibly participate in the communication with the expression platform. Our model provides molecular level explanations for existing experimental data, including thermodynamic and kinetic data, related to the ligand binding and switching process.

Abstract

Riboswitches are ligand dependent regulatory elements, found in mRNA transcripts, which are composed of two distinct, yet mutually interacting domains: a ‘ligand-binding’ or aptamer domain and an expression platform involved in gene regulation. Conformational changes due to ligand binding, controls downstream gene expressions. Purine riboswitches have been widely investigated for their remarkable ligand specificity and variation in functional mechanism, despite their sequence similarities. Yet, a complete molecular level understanding of the mechanism of ligand binding, ligand discrimination and switching is still limited because the characterization of the ligand-free OPEN state is limited primarily to results derived from chemical probing, NMR, fluorescence, single labeled FRET and single molecule force spectroscopy experiments. We have carried out ab initio quantum chemical studies on the binding pocket1 and studied explicit-solvent molecular dynamics simulations of the aptamer domain of the adenine riboswitch both in the adenine-bound (CLOSED) and the adenine-free (OPEN) state.2 Dynamic cross correlation, root mean square fluctuation and backbone torsion angle analyses of the trajectories and analysis of the average OPEN and CLOSED state structures, reveal alternative hydrogen bonding schemes leading to the switch and provides molecular level explanations for existing experimental data related to the ligand binding and switching process

Abstract

Unlike regular double stranded DNAs, which are stabilized through the formation of canonical Watson Crick base pairs, 3D structures of self-folded single stranded RNA molecules utilize a variety of noncanonical base pairs in order to attain their stable native geometry. Understanding the protein-like complex functionality of RNA molecules, at the molecular level, requires detailed studies on the geometries, interaction energies and stabilities of these base pairs in their respective structural and functional context of occurrence. Rapid growth in the number of solved RNA structures, and the availability of algorithms for automated detection and geometric classification of base pairs, have provided us with a challenging opening for quantum chemical inputs in the exciting area of RNA structural biology.[1-8] Here we specifically discuss noncanonical base pairs belonging to the Cis Hoogsteen-sugar edge (H:S) family, also known as the platform family, which play important structural role in folded RNA molecules.[1] We present results of detailed theoretical investigations of these base pairs, which involve sugar-base hydrogen bonding through 2-OH group of ribose. The studied structures were relaxed at the B3LYP/6-31G(d,p) level, and interaction energies were derived at the RIMP2/aug-cc-pVDZ level of theory. The geometries for each of the studied base pairs were characterized in terms of number and nature of H-bonds, RMSD values observed on optimization, base pair geometrical parameters and sugar pucker analysis. The rich variety of hydrogen bonding pattern, involving the flexible sugar edge, appears to hold the key to several features of structural motifs, such as planarity and propensity to participate in triplets, observed in this family of base pairs. Our studies explore these aspects by integrating database analysis, and detailed base pairing geometry analysis at the atomistic level, with ab-initio computation of interaction energies. In addition, alternative classification of base pairs and triplets, which provides insights into intrinsic properties of these base pairs and their possible structural and functional roles, will be discussed.

Abstract

Noncanonical base pairs are of key significance for the complex protein-like functionality of RNA molecules. Availability of a large number of solved crystal structures of functional RNA molecules, such as tRNAs, rRNAs and a variety of ribozymes and riboswitches, and that of an efficient scheme for geometric classification of base pairs, have vastly facilitated quantum chemical investigations in the exciting area of RNA structural biology. Here we report results of our combined quantum chemical and database analysis study on base pairs belonging to two of the twelve geometric families: the cis and trans H:H base pairs. We identify three previously unreported base pair combinations, two from cis H:H, and one from trans H:H family. We also sort out the relative order of strengths of interbase interaction, and correlate it with the observed occurrence frequencies of these base pairs. We find that some base pairs which are nonplanar in their isolated minimum energy geometries attain planarity and stability on triplet formation. The strongest and also the most frequently occurring among these, the A:A H:H trans base pair, occurs at many distinctly conserved positions in known rRNA and tRNA structures. Our quantum chemical studies reveal that though this base pair is intrinsically nonplanar, it has the tendency to attain stable planar geometry in RNA crystal structures through participation in higher order tertiary interactions including additional base-phosphate interactions. The tendency to form such additional interactions correlates well with their high occurrence frequency and is suggestive of their significant role in the context of RNA folding.

Abstract

Trans Hoogsteen/sugar edge (H/SE) RNA base pairs form one of the six families of RNA base pairs that utilize the 2-hydroxyl group of ribose for base pairing and play key roles in stabilizing folded RNA molecules. Here, we provide a detailed quantum chemical characterization of intrinsic structures and interaction energies of this base pair family, along with the evaluation of solvent screening effects by a continuum solvent approach. We report DFT-optimized geometries and MP2 interaction energies for all 10 crystallographically identified members of the family, for a representative set of them, using complete basis set extrapolation. For 6 of the 10 base pairs, we had to apply geometric constraints to keep the geometries relevant to RNA. We confirm that the remaining, hitherto undetected, possible members of this family do not have appropriate steric features required to establish stable base pairing in the trans H/SE fashion. The interaction patterns in the trans H/SE family are highly diverse, with gas-phase interaction energies in the range from 1 to 17 kcal/mol. Except for the C/rC and G/rG trans H/SE base pairs, the interaction energy is roughly evenly distributed between the HF and correlation components. Thus, in the trans H/SE base pairs, the relative importance of electron correlation is noticeably smaller than in the cis WC/SE or cis and trans SE/SE base pairs, but still larger than in canonical base pairs. The trans H/SE A/rG base pair is the intrinsically most stable member of this family. This base pair is also known as the sheared AG base pair and belongs to the most prominent set of RNA base pairs utilized in molecular building blocks of functional RNAs. For all trans H/SE base pairs that we identified, in addition to conventional base pairing, viable alternative structures were stabilized by amino-acceptor interactions. In the QM calculations, these amino-acceptor complexes appear to be equally as stable as those with common H-bonds, and more importantly, the switch to amino-acceptor interaction does not require any significant geometrical rearrangement of the base pairs. Such interactions are worthy of further investigations, as X-ray crystallography cannot unambiguously distinguish between conventional and amino-acceptor interactions involving the 2-hydroxyl group, formation of such interactions is usually not considered, and molecular modeling force fields do not include such interactions properly as a result of neglect of amino-group pyramidalization.

Abstract

Riboswitches are structural cis-acting genetic regulatory elements in 5' UTRs of mRNAs, consisting of an aptamer domain which regulates the behavior of an expression platform in response to its recognition of, and binding to, specific ligands. While our understanding of the ligand-bound structure of the aptamer domain of adenine riboswitches is, based on crystal structure X-Ray data and is well characterized, understanding of the structure and dynamics of the ligand free aptamer is limited to indirect inferences from physico-chemical probing experiments. Here we report results of 15ns long explicit-solvent molecular dynamics simulations of the add A-riboswitch crystal structure (1Y26), both in the adenine bound (CLOSED) state and in the adenine free (OPEN) state. Root mean square deviation (RMSD), root mean square fluctuation (RMSF), dynamic cross correlation (DCC) and backbone torsion angle analyses carried out on the two trajectories; and solvent accessible surface area (SASA) analysis of the two average structures, are benchmarked against available experimental data and are shown to constitute the basis for obtaining reliable insights into the molecular level details of the binding and switching mechanism. Our analysis reveals the interaction network responsible for, and conformational changes associated with, the communication between the binding pocket and the expression platform. It further highlights the significance of a, hither to unreported, non canonical W:H trans base pairing between A73 and A24, in the OPEN state, and also helps us to propose a possibly crucial role of U51 in the context of ligand binding and ligand discrimination.

Abstract

Schistosoma mansoni, a trematode parasite, which causes schistosomiasis and affects more than 200 million people worldwide, lives in an aerobic environment and therefore needs an effective redox mechanism for surviving reactive oxygen species from its host. Although, the host has two different redox systems: glutaredoxin and thioredoxin , the parasite has only one unique multifunctional enzyme, thioredoxin glutathione reductase (TGR) involving a fusion of two proteins, glutaredoxin (Grx) and thioredoxin reductase (TR), for performing all the redox activities. This dependence of S. mansoni on a single protein, TGR, for its protection from oxidative stress, makes it a promising drug target. Here, we describe a suitably validated, homology model for S. mansoni TGR (SmTGR), developed using both TR and Grx templates, functionally complete in the dimeric form with cofactors NADP(H) and FAD. Comparative analysis of substrate and inhibitor binding pockets of our model with crystal structures of parent TR as well as with that of glutathione reductase (GR), which is an essential component of the Grx system, appears to provide greater insight into the functioning of TGR. This also augments recent observations reported on the basis of X-ray structure data on SmTGR monomer lacking the C-terminal selenocysteine tail

Abstract

Graph theoretic concepts have been applied to the dynamic cross-correlation data obtained from MD simulation of adenine riboswitch, in absence and presence of adenine. Our hybrid approach combines community detection algorithms that support edge weights and cliques. The effect of variations in the values of nearest neighbors (NN) and correlation coefficient threshold (T) in the community detection algorithm have been applied to identify and filter out coincidental correlations between rogue nodes. Our results have identified the correlations within the structural regions of the molecule, which provide strong clues regarding the functionality and stability of the molecule in absence and presence of adenine. Our results show that a prior application of our algorithm applied to the simulation data of biomolecules in automated fashion can provide strong leads for hypothesis formulation and subsequent hypothesis-driven manual investigation.