Computational Design and Management in Pharmaceuticals

Computational design and management in pharmaceuticals

• Liu Sui

Abstract: Throughout the years since the computer was first developed, the computer has become required and indispensable in modern society. Significant scientific usage of computers has spread throughout all the sciences, including pharmaceutical science. In pharmaceuticals, usage has become an essential tool for the whole drug development process, from initiation of lead searching to finding the best fit, to finding toxicity. This paper will give an overview on how computers are used in the field of computational drug design.

Get Help With Your Essay

If you need assistance with writing your essay, our professional essay writing service is here to help!
Find out more about our Essay Writing Service

The development of computers is a short but exciting history. Looking back at this short history, it perfectly illustrates the intelligence and grittiness of mankind. Since the invention of electronic intelligence, this industry has been growing at an amazing pace. From the technical point of view, computers have changed a huge amount since ENIAC in 1946 to the modern day Intel and ARM architectures permeating our life (Bellis). Computers have changed their role from supercomputer being used for big companies and organizations to the personal computer that exists in just about every household, in one way or another. IT companies have also changed from marketing huge computers to marketing tiny computers to be used in the household, that synchronize with the fast speeds of the modern internet. In 1946, ENIAC was co-operated by the U.S. government and the University of Pennsylvania department of computer science (Goldstine). Features of this first generation of computers were that operating instructions were prepared for a specific task, and each machine had its own different mechanical languages. This generation of computers had very limited functionality and slow processing speed. Nonetheless, in less than 60 years, computers have become tools that are used by many different fields of study to enhance their overall value.

The rapid development of computer technology has led to a massive expansion of computer-related applications in the pharmaceutical industry. From the local computer system-based assistance, to the inevitable development of network-based assistance, usage of computer networks has become an inevitable trend. Both the computer industry and the pharmaceutical industry influence each other, and the combination of penetration, has and will continue to impact the operating mode of pharmacy. Management and technical decisions made ​​today within the pharmaceutical industry can be combined with the development of computer technology. All pharmacy workers should be aware of this and any future developments. In recent decades, due to the application of computers in pharmaceutical technology, many important achievements have been achieved.

Since antiquity, humans have built many tools to physically extend their physical capabilities such as the wheel, the pulley, and the vehicle. On comparison, creating devices to extending mental capability, such as the abacus, calculator, and computer, has also been a great human achievement. Computers are unlike any other tools in which they can replace human labor under pre-programed condition for an indefinite duration. An item only becomes useful for society depending on its function, where computers have many unique features to make them ideal for society: • Computers have incredible calculation ability. • Computers have a huge memory, in order to go through large data sets

The CPU and GPU of a computer have the ability to perform billions of complicated math operations per second. In terms of the pharmaceutical technology industry, this huge processing speed is vital for the complex mathematical operations required of this emerging discipline, such as calculating pharmacy finances, calculation and maintaining of pharmaceutical inventory, all the way to calculating drug and other protein formulas, determining the computational drug metabolism and its related pharmacokinetics computing, and pharmaceutical pattern recognition. Many computer-based programs have been developed and continue being improved to fill the huge needs of this industry. In the developmental stage of drug design, to search for drugs that possess the lowest energy in chemical structure can take a very long time, and is hard to do. Many people may question why do we need to calculate the lost energy possibility structure, and this question cannot be answered in one simple sentence.

In chemistry, each element is giving a symbol. Molecular formulas use these elemental symbols to show substantive (whether it’s an element or a molecular compound) composition of molecules and their relative molecular weight. Chemical formulas are widely used to present chemicals and chemical reaction. In nature, many drugs have different chemical formula, but at the same time, some compounds that have the exact same molecular formula may not be the same compounds; these compounds are called isomers. Isomers have same chemical formula but different atomic arrangement, and the cause of isomers is the change of order between atoms or groups. One type of isomers is called constitutional isomer. For example, ethanol and methoxymethane both have the chemical formula of C2H6O, but ethanol has an alcohol group, and methoxymethane has an ether function group (figure 1) Figure 1: chemical formula of ethanol and methoxymethane. This is where the software, Gaussian, becomes an invaluable part of pharmaceutical chemistry. Gaussian and its related software, Gaussview, are used to search for conformation amongst molecules. Stereoisomers have the molecular atom and group connected to each other in the same order but have different spatial arrangements. Many people may ask why we should care about the spatial arrangements. To answer this question, we need to think about molecules in terms of Classical Physics versus Quantum Physics. In chemistry, each chemical bond contains potential energy. The higher energy level the compounds bonds at, the less stable the compounds becomes. To make a more stable compound is a goal for many chemists because stable compounds have less of a chance to be decomposed, and in nature, many natural products being produced are those in their lowest energy states. The Classical Mechanics approach is mainly used for study of macroscopic objects in slow to stationary motion. Through studying the measureable movements through experiments, chemicals’ optimal nuclear positions can be found, and the lowest energy state can be found through graphing. However, in chemistry, all chemical bonds are in constant vibration and the ability to study the energy state of electron distribution is more useful for finding the lowest possible energy state. The optimal distribution of electron can be done by quantum mechanics. To think of energy as waves in the ocean, the lowest possible position is actually quite hard to find. It is possible to find some bumps, but these might not be the lowest points. To find the lowest energy points, huge amounts of calculation are needed. At this point, computers become essential. By inputting atomic coordinates, model chemistry and basis set into the software, Gaussian, the software will do the rest of the calculations and provide correct output including atomic coordinates, energy, and a wave function. The wave function can be further interpreted into molecular orbitals, partial charges, electrostatic potential surface, chemical shifts, bond orders, and spin densities. In order to find all these information, a high degree of accuracy is required. However, computers only have a certain amount of accuracy: they can only simulate continuous functions and numbers up to a finite point, leading to an accuracy problem. In general, most chemistry calculations have a certain degree of error that is allowed as long as the relative error in within the sufficient acceptable error range. Theoretically, the precision of calculation by computer is unlimited, but in practicality, most only go as far as a certain amount.

Beyond building a drug at its lowest energy state, or find a drug’s real conformation with incredible speed and accuracy, the huge data storage and memory capacity allow for huge amounts of library research. There are huge online drug repositories (both public and private) for researchers and scientists to search for their targeted drug. During the drug development phase, the first part of any research is to screen for lead compounds and modify these lead compounds to make them work on human biology. Because there are literally millions of compounds available to start from, how should one most efficiently find the compound desired? The answer is through computational lead compound search. Computers will input parameters given and search for lead compounds that fit the requirements and list them out with more information. Information retrieval of drug related data is an essential tool in the pharmacist’s tool belt.

One example of a great computational research tool used for computational design of drugs is the OpenEye OMEGA software suite. OMEGA is the name of a software product belonging to the OpenEye scientific software suite. OMEGA is a powerful tool for screening toxic chemical groups and providing validation of Lipinski’s rule of 5. OMEGA and vRocs have large libraries that can provide much help throughout the usage of computer-aided drug design. The OpenEye product claims that it “performs rapid conformational expansion of drug-like molecules, yielding a throughput of tens of thousands of compounds per day per processor (open eye website)”. This is a huge search, and without the modern memory, data storage, and speed of modern computers, this task would be impossible. At the beginning of any computational research, researchers have to get into a specific mindset. First, what disease does this researcher want to work on? Based on the disease selected, what drugs are currently on the market? Third, are there any other drugs can be any possible new drug candidates? At this point, researchers can start putting their desired pharmacophora into a computer, and let the computer search the library to suggest any possible candidates for further research.

Automated drug screening is a good example of this type of raw processing speed and breadth of data to go through. Extensive automatic pharmacological screening for compounds is the traditional and effective method to find new drugs. The sources of compounds are available for screening on a wide range of values including synthetic compounds, natural extracts, microbial fermentation, and compounds obtained by combinatorial chemistry techniques. There are a large number of these compounds possible, so in order to avoid leakage of data across screenings, screening needs to go through a few dozen general pharmacological screening models. To have the best possible outcomes, usually the combination of computer and robates for a netter system can run a screen quickly, efficiently, and on a large scale of samples. Currently, 10μg of a typical compound is a sufficient amount to go through dozens of pharmacological screenings, and as tens of thousands of compounds can be screened per day, this provides valuable research and development of lead compounds.

Within the past few years, even the regular computer is able to store a staggering amount of information. In order to perform the screening methods mentioned above, computers need to have large libraries. However, having a large library is not enough for computer to perform computational research; a certain amount of AI logic is also required. This AI logic ability as implemented through judgment causality analysis is the ability to analyze the proposition being established in order to make the appropriate countermeasures. This logic, or pattern recognition, is nowadays easily implemented by computers. OMEGA is one program can be used for pattern recognition.

Drugs are used to cure diseases, but for many drugs, they can be toxic to human at the same time as they are helping us control and cure some diseases. Pattern recognition uses the computer using mathematical methods to study automatic processing techniques and interpretation models. We consider the environment and objects within as a “model.” With the development of computer technology, it is possible to model extremely complex human information processing. An important form of this type of modeling is the identification process on the environment and the living body object. OMEGA can take as input information on the compounds generated by Gaussian and run through GaussView to filter out toxic compounds. This filter can recognize extremely complex pattern. In this filter, many structures are programed in as toxic groups. Any compounds possessing properties of any of these toxic groups will not pass this filter. Other than toxic groups, this program can also recognize number of hydrogen bond donors (HBD) and hydrogen bond acceptors (HBA). HBD and HBA counts are important for drugs because they are important indication for if a drug candidate can be a production drug or not. Dr. Lipinski is the scientist who first comes up with a so-called “rule of five.” Linpinski’s rule of five was created in 1997 after Christopher A. Lipinski studied 2245 drugs appear on the World Drug Index that have passed phase II clinical trials. By study these drugs’ structural features he came up with four rules:

1. The molecule weight of these compounds less than 500.
2. The number of HBD is less than 5.
3. The number of HBA is less than 10.
4. Log P is less than 5 (Lipinski) (Lipinski et al)

Because of Lipinski’s study, the number of HBA and HBD become one critical point when dealing with finding new drug candidates. The variable “P” is the lipo-hydro partition coefficient, and Log (P) is used to measure the solubility comparison of a compound’s solubility of octane to water. In order to pass through the body, drugs need to be polar in order to dissolve in the bloodstream. However, a drug should not be too polar, because it needs lipophilicity to pass through cell membrane. OMEGA is able to filter all these individual factors, and provide the end user a spreadsheet with all the information contained.

After initial candidates search, it is time to test if the drug has a good binding to the target protein. In the human body, drugs need to bind to target protein thereby either inhibit or excite a series of biological reactions. How well a drug can bind to its target directly affect this drug’s efficiency. This structure-activity relationship is related to a drug’s pharmacokinetics and pharmacodynamics. The chemical structure affects a drug’s properties, and these structures will decide which protein this drug will interact with. A drug should not be too tightly bound to the protein because in this case the drug will be very hard to be metabolized and eliminated through the body, and can cause accumulation in the body, and be toxic. To measure how well a drug can bind to its target, the software VIDA is the best choice. VIDA is a program which can visualize docking results of the drug with the protein in a 3D view.

Beyond this entire skillset of detailed programs within pharmaceutical chemistry, it is also nice to have a computer that is easy to use, able to perform automatic work, and bind all these programs together. As more and more modern drug analysis use computer instruments for analysis, so many different analytical instruments and computer connection and so many different instrumentation and automation for online use are not only for the determination of electrochemical, spectroscopic, kinetic equilibrium constant, but they are also used for data processing, statistical analysis and results. This all will allow for drug analysis continue to move forward in a sensitive, accurate and rapid direction.

Over the years, computer has been developing rapidly, and at the point, people are not only working on making computer faster. Instead, people trying to put this powerful Programs are designed for people, and by people, reflecting the people’s way of thinking and behavior of action, remember to replace part of the program and will be able to simulate human thinking and activities.

Reference

1. Bellis, Mary. “The History of the ENIAC Computer.” About.com Inventors. About.com, 16 May 2014. Web. 03 June 2014.
2. GOLDSTINE, HERMAN H. “Computers at the University of Pennsylvania’s Moore School, 1943-1946.” Computers at the University of Pennsylvania’s Moore School, 1943-1946. PROCEEDINGS OF THE AMERICAN PHILOSOPHICAL SOCIETY, 1992. Web. 04 June 2014.
3. Lipinski, Christopher A. “Lead- and Drug-like Compounds: The Rule-of-five Revolution.” Lead- and Drug-like Compounds: The Rule-of-five Revolution. Elsevier B.V., Dec. 2004. Web. 04 June 2014.
4. Lipinski, Christopher A., FRANCO Lambardo, Beryl W. Dominy, and Paul J. Feeney. “Experimental and Computational Approaches to Estimate Solubility and Permeability in Drug Discovery and Development Settings.” Experimental and Computational Approaches to Estimate Solubility and Permeability in Drug Discovery and Development Settings. Elsevier B.V., Mar. 2001. Web. 04 June 2014.

1