Pattern recognition and massively distributed computing

Davies, Elwyn; Glick, Meir; Harrison, Karl N.; Richards, William G.

doi:10.1002/jcc.10107

Cited by 20 publications

(9 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many alternative variations of these methods, such as neural network, discriminant analysis, cluster analysis, etc. [1,[3][4][5][6]. Most of the research results obtained by the specific methods describing various kinds of biological activity allow one to identify the molecular fragments most often found in drugs.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A new paradigm for pattern recognition of drugs

Потемкин

Grishina

2008

J Comput Aided Mol Des

View full text Add to dashboard Cite

A new paradigm is suggested for pattern recognition of drugs. The approach is based on the combined application of the 4D/3D quantitative structure-activity relationship (QSAR) algorithms BiS and ConGO. The first algorithm, BiS/MC (multiconformational), is used for the search for the conformers interacting with a receptor. The second algorithm, ConGO, has been suggested for the detailed study of the selected conformers' electron density and for the search for the electron structure fragments that determine the pharmacophore and antipharmacophore parts of the compounds. In this work we suggest using a new AlteQ method for the evaluation of the molecular electron density. AlteQ describes the experimental electron density (determined by low-temperature highly accurate X-ray analysis) much better than a number of quantum approaches. Herein this is shown using a comparison of the computed electron density with the results of highly accurate X-ray analysis. In the present study the desirability function is used for the first time for the analysis of the effects of the electron structure in the process of pattern recognition of active and inactive compounds. The suggested method for pattern recognition has been used for the investigation of various sets of compounds such as DNA-antimetabolites, fXa inhibitors, 5-HT(1A), and alpha(1)-AR receptors inhibitors. The pharmacophore and antipharmacophore fragments have been found in the electron structures of the compounds. It has been shown that the pattern recognition cross-validation quality for the datasets is unity.

show abstract

Section: Introductionmentioning

confidence: 99%

“…They permit compounds to be classified as active or inactive. It is possible to make a supposition about the binding mechanism of the described agents [e.g., 2,4,6]. However, the given approaches are usually used for the consideration of sets with a unique scaffold [e.g., 5,[8][9][10].…”

Section: Introductionmentioning

confidence: 99%

A new paradigm for pattern recognition of drugs

Потемкин

Grishina

2008

J Comput Aided Mol Des

View full text Add to dashboard Cite

show abstract

“…The Folding@Home from Stanford University is designed to understand protein folding and related diseases. The "Screensaver Lifesaver" projects from research group of W. Graham Richards has developed and ap-plied computational methods for drug design, molecular similarity analysis and protein structure prediction, and performed simulations of enzyme reaction mechanisms, DNA recognition, and lipid bilayers [9] [10].…”

Section: Idea@home: Intelligent Data Engineering Platformmentioning

confidence: 99%

Genetic Mining of DNA Sequence Structures for Effective Classification of the Risk Types of Human Papillomavirus (HPV)

Eom

Park

Zhang

2004

Neural Information Processing

View full text Add to dashboard Cite

Abstract. Human papillomavirus (HPV) is considered to be the most common sexually transmitted disease and the infection of HPV is known as the major factor for cervical cancer. There are more than 100 types in HPV and each HPV has two risk types, low and high. In particular, high risk type HPV is known to the most important factors in medical judgment. Thus, the classifying the risk type of HPV is very important to the treat of cervical cancer. In this paper, we present a machine learning approach to mine the structure of HPV DNA sequence for effective classification of the HPV risk types. We learn the most informative subsequence segment sets and its weights with genetic algorithm to classify the risk types of each HPV. To resolve the problem of computational complexity of genetic algorithm we use distributed intelligent data engineering platform based on active grid concept called "IDEA@Home." The proposed genetic mining method, with the described platform, shows about 85.6% classification accuracy with relatively fast mining speed.

show abstract

“…In this project, a screensaver is loaded on to local PC's to perform most of the compute-intensive docking [105]. This project derives from the well-known SETI@home (Search for Extraterrestrial Intelligence) project [106].…”

Section: Docking and Screeningmentioning

confidence: 99%

Ligand-Protein Docking: Cancer Research at the Interface between Biology and Chemistry

Rc¹,

2003

CMC

View full text Add to dashboard Cite

In recent years there has been a growing interest in computer-based screening. One of the driving forces has been the increased efficiency of protein crystallography leading to the real possibility of using structure-based design as a significant contributor to the discovery of novel ligands. In 1957 after 22 years of work the first protein structure, determined by x-ray crystallography was produced. Now the process has become increasingly automated and nearly 20,000 protein structures are available in the Protein Data Bank (PDB). Equally, progress in genomics will result in a great expansion of validated targets for cancer therapy. The understanding of the relationships between structure and function of gene products will be one of the key routes to new therapeutic advances. The challenge now is to use this data in the discovery of novel therapeutics. One approach is obviously to synthesize molecules and co-crystallize or soak them into the protein crystal and so determine the position and interaction of the molecule with the protein. The structural information obtained (where does the molecule bind; what are the ligand/protein/solvent interactions?) can be invaluable in the generation of novel molecules or in the re-design of existing molecules whose drug properties are not optimal. However, when dealing with large numbers (millions) of molecules, when crystallization is difficult or in testing hypotheses, a significant contribution can be made using computer based screening methods. In order to use the structural information derived from x-ray crystallography (or other sources, for example NMR or homology modelling) when evaluating the utility of a novel ligand, we need to understand where in the protein (or other macromolecule such as RNA) the ligand is likely to bind and also if possible, the strength of the binding interactions. This problem is known as the 'docking problem'. There have been many approaches to the solution of this problem over the last ten years. For example, some methods rely on complex molecular dynamics simulations while others use less costly graph matching approaches. There is generally a compromise between speed and accuracy, with some methods giving much more information and insight into the nature of the protein/ligand interactions and other methods optimised for speed of docking thousands of putative ligands. We will describe some of the more common methods and algorithms used to solve the docking problem and in particular, we will review recent applications in cancer research.

show abstract

Pattern recognition and massively distributed computing

Cited by 20 publications

References 41 publications

A new paradigm for pattern recognition of drugs

A new paradigm for pattern recognition of drugs

Genetic Mining of DNA Sequence Structures for Effective Classification of the Risk Types of Human Papillomavirus (HPV)

Ligand-Protein Docking: Cancer Research at the Interface between Biology and Chemistry

Contact Info

Product

Resources

About