Man Hon Wong

iview: an interactive WebGL visualizer for protein-ligand complex

BackgroundVisualization of protein-ligand complex plays an important role in elaborating protein-ligand interactions and aiding novel drug design. Most existing web visualizers either rely on slow software rendering, or lack virtual reality support. The vital feature of macromolecular surface construction is also unavailable.ResultsWe have developed iview, an easy-to-use interactive WebGL visualizer of protein-ligand complex. It exploits hardware acceleration rather than software rendering. It features three special effects in virtual reality settings, namely anaglyph, parallax barrier and oculus rift, resulting in visually appealing identification of intermolecular interactions. It supports four surface representations including Van der Waals surface, solvent excluded surface, solvent accessible surface and molecular surface. Moreover, based on the feature-rich version of iview, we have also developed a neat and tailor-made version specifically for our istar web platform for protein-ligand docking purpose. This demonstrates the excellent portability of iview.ConclusionsUsing innovative 3D techniques, we provide a user friendly visualizer that is not intended to compete with professional visualizers, but to enable easy accessibility and platform independence.

An expert-system shell using structured knowledge: an object-oriented approach

An architecture for an expert-system shell that mixes declarative nd procedural knowledge, overcoming a major problem of conventional shells, is presented. The prototype shell uses structured knowledge representations and its built-in database interface not only allows automatic extraction of data from a database management system but also provides a fuzzy database query facility. The shells object-oriented approach to knowledge representation supports data and knowledge acquisition and management. Another feature is encapsulation which prevents object manipulation except by defined operations. A comparison of representation methods and two case studies showing System X-Is power and flexibility are included.<<ETX>>

A fuzzy database-query language

Abstract Selection conditions for conventional database-query languages are not natural enough to express criteria with fuzzy concepts. To choose some alternatives from a database requires the techniques of multicriteria decision making or conflict resolution. However, some traditional techniques in this area still have some drawbacks, such as inefficiency and independency of criteria. In this paper a fuzzy database-query language is presented and its position in the area of multicriteria decisions making in database applications is discussed and identified. The selection constraints of this query language can accept both crispy and fuzzy conditions.

A fuzzy expert database system

Abstract Fuzzy concepts always exist in much of human reasoning as well as decision making. This paper presents a fuzzy expert database system which is an integration of a fuzzy expert system building tool called SYSTEM Z-II and a database management system called Rdb/VMS. This system is able to extract fuzzy data and terms stored in a database and used in the fuzzy reasoning in an expert system. It can also retrieve information by fuzzy database-queries which are generated by the expert system automatically. Many expert systems in different domain areas such as decision making can be constructed. Sample applications are described to demonstrate the flexibility and power of this system. The fuzzy query language defined and used in the system can also be used independently as a fuzzy enquiry tool in database applications.

Substituting random forest for multiple linear regression improves binding affinity prediction of scoring functions: Cyscore as a case study

BackgroundState-of-the-art protein-ligand docking methods are generally limited by the traditionally low accuracy of their scoring functions, which are used to predict binding affinity and thus vital for discriminating between active and inactive compounds. Despite intensive research over the years, classical scoring functions have reached a plateau in their predictive performance. These assume a predetermined additive functional form for some sophisticated numerical features, and use standard multivariate linear regression (MLR) on experimental data to derive the coefficients.ResultsIn this study we show that such a simple functional form is detrimental for the prediction performance of a scoring function, and replacing linear regression by machine learning techniques like random forest (RF) can improve prediction performance. We investigate the conditions of applying RF under various contexts and find that given sufficient training samples RF manages to comprehensively capture the non-linearity between structural features and measured binding affinities. Incorporating more structural features and training with more samples can both boost RF performance. In addition, we analyze the importance of structural features to binding affinity prediction using the RF variable importance tool. Lastly, we use Cyscore, a top performing empirical scoring function, as a baseline for comparison study.ConclusionsMachine-learning scoring functions are fundamentally different from classical scoring functions because the former circumvents the fixed functional form relating structural features with binding affinities. RF, but not MLR, can effectively exploit more structural features and more training samples, leading to higher prediction performance. The future availability of more X-ray crystal structures will further widen the performance gap between RF-based and MLR-based scoring functions. This further stresses the importance of substituting RF for MLR in scoring function development.

A fast projection algorithm for sequence data searching

Abstract In real life, data often appear in the form of sequences and this form of data is called sequence data. In this paper, a new definition on sequence similarity and a novel algorithm, Projection Algorithm, for sequence data searching are proposed. This algorithm is not required to access every datum in a sequence database. However, it guarantees that no qualified subsequence is falsely rejected. Moreover, the projection algorithm can be extended to match subsequences with different scales. With careful selection of parameters, most of the similar subsequences with different scales can be retrieved. We also show by experiments that the proposed algorithm can outperform the traditional sequential searching algorithm up to 96 times in terms of speed up.

Fuzzy concepts in an object oriented expert system shell

Fuzzy logic is one of the methods to model the vagueness and imprecision of human knowledge. Some rule‐based expert system shells have been successfully developed and have demonstrated the power of fuzzy logic in dealing with inexact reasoning and rule inferences. However, using rules for knowledge representation is not structured enough. In addition, knowledge cannot be easily represented in an abstracted (hierarchical) from. In this article the introduction of fuzzy concepts into object oriented knowledge representation (OOKR), which is a structured knowledge representation scheme, is presented. A framework for handling all the possible fuzzy concepts in OOKR at both the dynamic and static levels is proposed. In order to handle the inheritance mechanism and to model the relations among classes, instances, and attributes, some new fuzzy concepts and operations are introduced. These concepts and operations are developed from the semantic meaning rather than by an ad hoc approach. A prototype of the expert system shell. System FX‐I, has been successfully developed based on the above framework, showing the feasibility of handling inexact knowledge in a structural way.

idock: A multithreaded virtual screening tool for flexible ligand docking

AutoDock Vina is a competitive protein-ligand docking tool well known for its fast execution and high accuracy. Nevertheless, when docking a massive number of ligands, Vina has to be run multiple times, repeating receptor parsing and grid maps building over and over again. There are tremendous requests for revising Vina to reuse precalculated data and incorporate built-in support for virtual screening. Hence we developed idock, inheriting from AutoDock Vina the accurate scoring function and the efficient optimization algorithm, and significantly improving the fundamental implementation and numerical model for even faster execution. idock achieves a speedup of 3.3 in terms of CPU time and a speedup of 7.5 in terms of elapsed time on average. idock is free and open source, available at https://GitHub.com/HongjianLi/idock.

Generalizing and learning protein-DNA binding sequence representations by an evolutionary algorithm

Protein-DNA bindings are essential activities. Understanding them forms the basis for further deciphering of biological and genetic systems. In particular, the protein-DNA bindings between transcription factors (TFs) and transcription factor binding sites (TFBSs) play a central role in gene transcription. Comprehensive TF-TFBS binding sequence pairs have been found in a recent study. However, they are in one-to-one mappings which cannot fully reflect the many-to-many mappings within the bindings. An evolutionary algorithm is proposed to learn generalized representations (many-to-many mappings) from the TF-TFBS binding sequence pairs (one-to-one mappings). The generalized pairs are shown to be more meaningful than the original TF-TFBS binding sequence pairs. Some representative examples have been analyzed in this study. In particular, it shows that the TF-TFBS binding sequence pairs are not presumably in one-to-one mappings. They can also exhibit many-to-many mappings. The proposed method can help us extract such many-to-many information from the one-to-one TF-TFBS binding sequence pairs found in the previous study, providing further knowledge in understanding the bindings between TFs and TFBSs.

Protein structure prediction on a lattice model via multimodal optimization techniques

This paper considers the protein structure prediction problem as a multimodal optimization problem. In particular, de novo protein structure prediction problems on the 3D Hydrophobic-Polar (HP) lattice model are tackled by evolutionary algorithms using multimodal optimization techniques. In addition, a new mutation approach and performance metric are proposed for the problem. The experimental results indicate that the proposed algorithms are more effective than the state-of-the-arts algorithms, even though they are simple.