Drug Designing with the Aid of Computers - Essay Example

? of Faculty of …………… Drug Designing with the Aid of Computers 24 Oct The advent of Computershas assisted the human society by innumerable means. Transforming the society into something that it was never like before, computers have become an integral part of our society. They have given the meaning to life that did not exist prior to them. This was mainly because computing devices possess extremely fast calculating capabilities and imaging mechanisms as a result of which analysis of data has now become possible that could not have even been thought of before the advent of computers. Drug Designing is an exceptional Example of a field that has been benefited immensely by the use of computers. Molecular imaging has made great revolutions in the designing of drugs. DRUG DESIGNING: Drug Designing can be termed as a method whereby drugs can be invented by targeting a specific biological molecule and then designing new medications based on the relevant information. If narrated simply, the designing of a drug may involve the following generic processes: The cause of the disease to be identified The target to be attacked is identified The ‘Lead Compound’ is identified. Tests are carried out for the ‘Lead Compound’ Trial conducted in clinics Drug approved Present day drug design extensively relies on computer based molecular modeling techniques. Computer Drug Design has been known to have possessed exemplary technologies that conduct the drug development process speedily. Computer Based Drug Designing Methodologies are preferred over Traditional Drug Designing capabilities because: TRADITIONAL DRUG DESIGNING is: “Random trial and erroneous Time consuming, Comparatively much more expensive Very low in output (1 in 100,000)” (Huang, n.d.) In Comparison to the above mentioned properties of traditional drug design Computer Aided Drug Design Methodologies are thought to be: Structural in Nature Specifically Targeted Automatic Much Faster than the Traditional Methods Much lower in cost Having a comparatively much higher success rate than the traditional method. DESIGNING A DRUG VIA COMPUTERS Molecular modeling is a popular method of drug design. It is a science by means of which molecular structures are represented numerically. Their behavior is then simulated by making use of the equations from quantum and classical physics. It is with the use of this molecular modeling via computers that computer aided drug designing finally gets executed. The first step in the approach towards drug design by this methodology is the identification of a molecule that may be critical to the pathogen of any specific disease. The molecular structure of the target molecule that has been identified is now resolute. Sufficient molecular information is essential in order to proceed further on with the designing of the drug. The structure of the target molecule, as discussed above, is used to construct a ligand. The recognition of the target protein is the next step followed by the ways by which it would be bound. The point of concern here is that the structure of proteins is very versatile in nature. Proteins change the structure of their molecule when they are undergoing different functions. Types of Molecular Modeling “Based on the type of information at hand, either ligand based or molecular based design methods are selected.” (Daharwal, 2006). The proposed combinations are very rapidly generated by computers that instantly display an entire list of the proposed molecules. Molecular Modelling screens through large databases of chemical ligands and molecules of compounds. In doing so, it identifies the molecules that may become potential drugs. The above process is termed as Molecular Docking. SIMULATIONS FOR DRUG DESIGNING Since protein structures are rapidly changing and cannot be predicted computers are essentially used to simulate the dynamically changing structure of molecules. Hihg performance computing is often used to assist. For example, at the company of NOVO Nordisk (McGrawHill, 2010) scientists have accepted high performance computing as a tool that would enable the study of complex molecular interactions. This was critically required for the recognition of ligands and the target proteins that were associated with them. The Initial Failure of CADD. The scientists were really optimistic about the prospective success of Computer Aided/Assisted Drug Design in the early 90s. They were literally expecting a revolution in the field of drug development. With a vertical/exponential trend exhibited by the computing power and real time graphical modeling of molecular structures the hopes were really high that the CADD would generate a considerable milestone in the basic approximations of ligand-receptor complementarities. The chemists started to believe that computational chemistry would considerably reduce the time and efforts involved in drug fusion and analysis by generation of new compounds through computational modeling. De Novo Design was the term designated for the method of creating unreal/virtual lead compounds wholly by computers. A hefty amount in dollars was spent by the world’s leading pharmaceutical firms in order to achieve a real world conversion of de novo design compounds. The new hardware and software were designed to accomplish the said goal but very unfortunately there were a very few cases where the computational techniques were proved successful. Overall de novo design was proved to be a great failure. The logical reasons behind this failure are explained in the following text. Assumptions and Approximations The main deficiency of the computational procedures is approximation of results due to the unavailability of high precision calculation capability. The approximated results always contain considerable errors which in the case of creating original lead compounds were proved to be the main hindrances. Optimization of algorithmic complexities Moreover the efficiency of algorithmic procedures is measured in terms of time and storage space involved in computation. The optimization attempts to improvise time complexities started assuring production of outputs, but the results were not up to the marks as the due brute force iterations were not in place. The precision of the interactions of any ligand-receptor was highly compromised due to the application of the above mentioned approaches. In many cases, several chemical structures that were already hypothesized by the chemists with a fair expectation of potential matching had proved to be having no correlation in reality in terms of calculated binding. Despite enormous computational power and speed the main challenge that was faced by the application of de novo design was that there were considerable approximations needed in order to forecast the ligand-receptor binding in a useful and considerable time frame. . The main objective for which the de novo design was destined was to dock into their receptors the ligands that were wholly generated from scratch. With this there was no assurance that how the estimated structure would act in reality? The creation of undesired chemical structures was an added problem with the computer assisted de novo design approach. The Success story The techniques of combinatorial chemistry and mass screening came to refuge for the survival of the de novo design technique. These techniques are used to explore and discover the lead compounds efficiently, quickly and in an authentic manner. The de novo design tools are further used to optimize the results of these techniques to produce powerful drugs. Drug optimization, in comparison to de novo design, constitutes a very important fact. The denovo ligand generation mechanism constitutes of the creation of an entire structure from scratch. The beginning of a drug optimization mechanism, however involves a lead compound. X-Ray crystallography is used to characterize the bound structure of the lead compound with the receptor. The derivatives of chemicals however are easily structured via computer automated mechanisms. The rapid binding and generation of all potential outcomes of the combinations can very well be predicted and generated by computers. Thus the basic idea is to improve or refine the weakly binding lead compounds by generating and predicting the binding of all possible derivatives to create a list of best candidates. Rational Drug Design Software There are three major categories of the software programs used for rational drug design. They are, Scanners Builders Hybrids. Scanners are generally used for the screening of lead compounds. Formerly designed databases are searched for the purpose. The matching structures as per pharmacophore are retrieved from the databases to generate a chance of complementing the receptor. The basic advantages of using these software packages are, The database contains known compounds so the issue of synthetic feasibility does not arise. Due to the usage of pre-formed databases the results are quick. The query results ensure matching compounds so the verification functions like scoring are not needed. The structural representations stored in databases proved to be authentic for small molecules. The limitations of these software packages are as follows, Having finite number of structures stored in databases there were limitations and chances of biases. The maintenance of these databases requires cost and personnel. Lack of these resources does not allow most of the educational institutions to maintain their databases. The range of possible hits is restricted as well. Builders are de novo design based programs. These are mostly used in optimization of the determined lead compounds. The advantages of builder programs are as follows: Database of predetermined structures is not required, so these programs can be used independently by anyone. The database that is required for the processing is built by the software itself. A variety of possible derivative structures can be attained through the addition of fragments in a combinatorial manner. A varied set of solutions is generated automatically that assists in generation of original ligands. The results of mass screening can be optimized by using these software packages. The disadvantages of builders are as follows, A large number of unfeasible structures is possible. The production of some chemically unstable ligands is proved. The software methodology governs the ways of attachments of ligands, which limits the diversity. Hybrids, as the name explains use combined approach. These programs mostly used de novo design method for the creation of lead compounds. These software packages are good in generating a large set of varied potential hits. However, these packages possess all the deficiencies of the de novo design methodology. Artificial Neural Networks in Drug Designing The ANN is a branch of artificial intelligence that simulates the functioning of the biological networks of neurons in human brain for decision making. This is the latest approach applied in the field of drug designing and molecular docking. This technique makes the computers learn as humans learn. The computers learn the ways to achieve a target molecule and learn to improve the results on their own. The complex calculations can be solved through this approach on merit. The associated weights to the inputs assure the accuracy of results to a greater extent. Drug optimization, in comparison to de novo design, constitutes a very important fact. The denovo ligand generation mechanism constitutes of the creation of an entire structure from scratch. The beginning of a drug optimization mechanism, however involves a lead compound. X-Ray crystallography is used to characterize the bound structure of the lead compound with the receptor. The derivatives of chemicals however are easily structured via computer automated mechanisms. The rapid binding a generation of all potential outcomes of the combinations can very well be predicted and generated by computers. Thus a list of the best potential candidates is created. The use of the computer aided software thus enables the refinement of compounds that have weakly bound the lead compounds in the most effective of manners. CAN CADD SUBSTITUTE TRADITIONAL DRUG DESIGNING Proteins are getting unfolded with every passing minute. The primary sequences of proteins represent their initial data. The methods of acquiring the secondary or tertiary structures of proteins are still not mature. This is the main reason due to which the structures of most of the receptors and ligands remain unknown. Since the proteins are unpredictable structural compounds as yet, therefore the accuracy of drug designing can not wholly be put on computer aided drug designing. Thus, it has often been found that the resultant final molecular drug structures of computer aided software have not proven their mettle when they are tested in the testing phase. Computer based drug design can greatly be used to assist the traditional drug design methods by reducing the set of possible matches to the few best and most probable matches. This benefits the cause by reduction of procedural cost, time and effort to a great extent. REFERENCES: Daharwal, J. S. 7 December 2006. Computer Aided Drug Design and BioInformatics: A current tool for Designing. Viewed 24 October, 2011 < http://www.pharmainfo.net/reviews/computer-aided-drug-design-and-bioinformatics-current-tool-designing > Bartusiak, M. 1981. Designing Drugs with computers. Viewed on: 24 October 2011. http://www.marciabartusiak.com/uploads/8/5/8/9/8589314/designing_drugs.pdf  Leach, R. A. 2010. Molecular Modeling Principles and Applications. Pearson. South Asia. McGraw-Hill. 2010. Computer-Aided Drug Design. viewed, 24 October 2011. < http://www.accessengineeringlibrary.com/mghpdf/0071764275_ar001.pdf > Dean, P. M. Lewis, Richard A. 10/1999  Molecular Diversity in Drug Design. Kluwer Academic Publishers Hingham, MA, USA  Read More