Full Text Article

Introduction

We aim to achieve this review paper by explaining AI's role in chemistry specifically in drug design and molecular modeling. If we want to describe AI using computer intelligence in the prediction structure of molecules and atoms in the chemical organic structure it can help the human mind potentially design and create chemical structures for predicting and analyzing drugs for the treatment of human diseases that are not cured with possible drugs [1]. In the pharmaceutical industry, AI can reach vast success in drug design research, this is because using AI technology can help scientists with the use of novel methods can design new drugs for treatment persons in the world and can reduce human manual error in designing drugs from structures of chemical complicate compound and it can help to hopeful landscape in future in pharmacology science in the world [2]. One another important key that we can apply to it is the selection of proper compounds in drug design strategies. The compound that is selected for more study should well physiochemical properties and ADMET standard items, all this process is done by using AI and machine learning in the pharmacy industry because AI can represent vast chemical information for choosing profit compounds in the drug design approach for chemists [3]. If we want to introduce AI can say it is a machine that constitutes human and can do every work such as humans and it may be applied in vast works education, study, research, lifestyle, and … AI is computer-based human intelligence and can be the error that human with higher mind potent can restrict activity it in some cases [4]. Another term that I intend to explain is organic chemistry means compounds that have non-aquatic reactions they can solve in compounds that have similar specificity to them. Furthermore, organic chemistry consists of all compounds that have in their structure carbon atom and non-organic chemistry includes all atoms without carbon [5]. At this time there are very need to use AI in creating drugs for patients. In the past delivery of drugs to the individual was a very time-consuming and complicated process that AI scientists could use with faster activity these drugs create for individuals that require them, also patients should be trained in consumer from this novel technology that may be long for 3 or 4 years [6]. Studies that down in recent years show that producing new drugs can be time-long, expensive, and very scarce in availability to patients with the help of data from AI and machine learning can be this issue solved but it needs more try and study in the future [7]. AI technique can integrate the help of machine learning from structures in organic chemistry to predict and propel new drugs that can be very useful in the therapy of rare disorders but are applicable from AI with limitations such as information, availability, and error, in this subject requires more clear options [8]. This review paper intends to research using collection information from other manuscripts in AI technology applied in the process design novel drugs from evaluating chemical structures for the treatment of different disorders. The main goal of this paper is to help variant AI approaches and machine learning strategies compare the differences and similarities of these methods can explain these results and reach a proper conclusion in designing novel drugs for individuals with different conditions. One research question on this topic is to evaluate the proper and non-proper effects of AI applied in drug design and molecular modeling in this process and that reply to this question is very hard and requires more clearing issues and research in this discipline. translating information from AI language to design novel drugs can play an important role in retaining the health of patients all over the world. As told later in the above text the creation of one drug for a specific disease is a very complex, long-time, and high cost in the world this use of innovative technology can be more helpful to scientists and patients in getting their health. Therefore, we can compare the main ideas of other authors to understand which method is the best choice for scientists to save time, is cheaper, and is more effective in creating drugs for diseases. One of the most important and routine programs used for the application of AI in organic chemistry is the SYNCHEM program. This approach system helps chemists resolve complicated chemical structures that with user handling, work is very hard for scientists specifically in stereochemistry content [11]. Scientists in organic chemistry in the use of AI in organic chemistry can achieve this success with the use of an axiomatizing program that can resolve optimum synthetic chemical organic reactions that chemists with only cannot find the best resolve method to proceed them [12]. One of the novel approaches that scientists use of AI in organic chemistry is using of novel logic languages for resolving and optimizing reactions chemical that easy methods cannot find, this novel method is done by a Prolog planner in the organic chemistry discipline [13]. One of the challenges in organic chemistry is designing and creating novel drugs for different conditions. Many chemists cannot chemical structures in organic chemistry to predict and design new drugs because it is time-consuming and high-cost to perform experimental work that uses artificial intelligence approaches to help chemist researchers synthesize new and novel drugs for rare fatal diseases all over the world [14]. In the new center, Scientists can use AI tools and organic chemistry software for the fast production of drugs that are very limited in different regions of the world and they successfully could teach this algorithm to students in academic universities and teaching places that this successful goal in near future is provided soon [15]. One of the methods in drug design is AI technology can help researchers with different machine learning to easily analyze chemical data and apply them to create novel drugs that can help patients in the world reduce their disease furthermore, AI technology can reach scientists with more accuracy and efficiency predict behavior of different drugs for various disease [16]. Table (1) summaries results from the use of AI in optimization chemical reactions:

Research Material

We can use in this review manuscript different approaches of machine learning in the direction of better use of AI in drug design with the help of organic or inorganic chemistry structures. Machine learning is the very best technique that can analyze information from AI faster in creating different drugs with vast organic chemistry structures that in condition usually this work is backbreaking and time-consuming. For example, the eToxPred model of machine learning can easily evaluate the amount of toxicity of small organic chemistry molecules with high efficiency. One of the methods evaluated in this paper is using of the QSAR model helps scientists interpret the quantitative specificity of organic molecules in chemistry science, also the method can help in predicting biological activity from the chemical structure which can be very useful in this project.

Main Body

In a study done by pu and et all in 2019, they used from eToxPred model of machine learning to assess the amount of toxicity of their chemical agents the result shows that this method can be useful in virtual screening and can be a good matter for evaluating the toxicity of chemical compounds [9], while another study by Ai and et all in 2019 express that this model has accuracy, sensitivity, and specificity for evaluate toxicity chemical agents in an aquatic system that as implicate in above text first case was good in virtual screening and the second model is the best for aquatic system [10]. From the interview of these two models, we can conclude that the eToxPred model only explores toxicity in the virtual system, and in another method this specificity ass in aquatic systems, and also both two matters are proper but notice which condition is ready scientists in chemistry science can apply these approaches for evaluate and fasting their results in their research. Furthermore, in these two models, there are other machine-learning strategies in vast studies of organic chemistry explained by authors that we implicate in this section shortly. One other method is supervised learning which evaluates the properties of the compound chemical from possible information, scientists say that this matter is the best and most useful when we have elementary data from one chemical structure and we don’t have the properties of this agent. This matter has 3 algorithms: 1. support vector machines: that from structure chemical information predict chemical properties of compounds that scare chemists. 2. Random forests: this method anticipates the properties of chemical compounds with the help of tree design in their matter which also causes high accuracy and both useful for complicated chemical structures. 3: Deep neural networks that is useful for predicting molecular properties of chemical compounds. Another approach is unsupervised learning this method is best when chemical information is scarce and with the help of sub-methods can predict chemical information. 1. Clustering algorithm: by visualization of chemical structures in space can reach this data and 2. Principal component analysis: useful for solving complicated chemical structures. Later model machine learning is transfer learning that is used when data chemical is limited and can be evaluated chemical properties from this scarce information, and the end approach is reinforcement learning that is useful for predicting efficiency conditions in chemical reactions. Scientists in chemistry can help these types of machine learning easily evaluate and predict chemical structures that can be important key points in innovation strategy in chemistry science.

Conclusion

This qualitative review has evaluated the effect of using AI technology in organic chemistry with the help of applying different types of machine learning approaches. In this review paper, we can reach these outcomes that each matter in its situation can be helpful and belong to a specific chemical structure and reaction with unique properties. But until now there have been limited studies that explain this concept in their paper and a few studies carried out on this topic. To clarify this issue in chemistry science young researchers should try to provide vast information about the use of each method in machine learning and applied AI technology in organic chemistry until now there are very gaps that should be filled to solve chemical issues in structure and performance for molecular modeling and design novel drugs in de novo approach for enhance accuracy, efficiency and short- time produce drugs technology. In Table 2 applicability of these AI approaches in organic chemistry is shown.