Smart Pharma: turn "big data" into "big data", search for promising compounds efficiently

Author: since the development of Wang Chan's biomedical industry, new technologies have been emerging As far as drug screening is concerned, it has experienced many changes, such as Shennong tasting herbs, manual screening, high-throughput screening and virtual screening However, the cruel reality is that the cost of new drug research and development is increasing year by year, and drug researchers are calling for more and more new technologies With the gradual landing of artificial intelligence in other fields, drug screening has become one of the new areas of artificial intelligence comprehensive application Dr Xie Weidong, CEO of smart Pharma, has been engaged in academic work in immunology and molecular biology in many famous universities in Canada and the United States After returning to China, he joined many famous Chinese pharmaceutical companies to hold positions of technical research and development, clinical transformation, etc in the pharmaceutical field He has more than 30 years of experience in the biomedical industry Dr Xie, who has been working in the field of biomedicine for many years, has a profound and unique understanding of drug screening He said: "20 years ago, we couldn't conceive" big data medicine ", because our medical data and computer computing ability are not ready Now, with the development of science and technology, the accumulation of medical data and the accessibility of hardware equipment (supercomputer) are ready With the soil of gestation, we smart medicine are committed to turning" big data "into" big data " Data "to make AI + new drugs really feasible." The original intention of smart Pharma is to use big data and algorithms to enable drug research and development, improve drug screening accuracy, shorten research and development time, and reduce research and development cost At present, the cost of drug research and development is huge Firstly, it takes too long for some drugs to appear Secondly, the cost of research and development is too high The average cost of compounds from research and development to the marketing of finished drugs is billions of dollars In addition, the process of drug research and development is also very difficult To select a compound with potential inhibition effect, pharmaceutical companies need to further verify its physicochemical properties, in vivo distribution, absorption, metabolism and toxicity If one of the factors is not up to the standard, it is necessary to reconstruct the drug Up to now, it is more and more difficult to obtain the promising compounds It is a time-consuming and labor-consuming process to screen natural products and compounds by random methods such as manual screening and high-throughput screening The quality of candidate drugs often determines the success or failure of follow-up preclinical and clinical development Ideally, in the initial stage of drug development, judgment and prediction of the physical and chemical properties, pharmacokinetics, toxicity and side effects of the lead compounds / candidate drugs will be made, so as to reduce the workload of drug development, shorten the research and development cycle, save research costs, and improve the success rate of development The emergence of virtual molecular docking makes it possible to realize the "ideal situation": to establish the geometric position matching function and energy function between small molecules and protein targets, to predict the interaction between small molecules and proteins, and to screen the seedling head compounds combined with protein targets on this basis Smart Pharma has collected the basic data of drugs on the market, new drugs under research, synthetic compounds and other drugs at home and abroad, as well as the protein data of various databases, research literature, patents and other accessible materials, and has established a number of different types of databases: nearly 100 million compound 3D structure databases and 130000 protein 3D structure databases, aiming at specific targets, from high to low From active to inactive, the database of compounds with different activity gradients was established After the completion of the early database construction, the company independently developed a drug screening model based on the compound molecular data and target protein data Through the deep convolution neural network model (CNN), the complex concepts are layered into a large number of simple local features, and a variety of feature extraction technologies are used to effectively extract structural features When analyzing protein structure, the model can analyze millions of sites in a three-dimensional structure or millions of parameters in a model at the same time The system can effectively collect multi-dimensional data, quickly and comprehensively analyze various index parameters, accelerate the progress of drug research and development, and improve the accuracy of prediction In several scenarios of AI enabled medicine, smart medicine focuses on the mining of small molecular compounds The company's self-developed deep learning algorithm system can establish drug screening model based on compound molecular data and target protein data The algorithm system can optimize the structure and function of the compound according to the target protein For specific proteins, the new drug screening system can design compounds with specific functions In addition, smart Pharma's drug discovery platform supports multi-user simultaneous operation, and can access data results from multiple ports The platform integrates a variety of artificial intelligence algorithms, and forms a multi-dimensional evaluation molecular protein combination screening scoring system, integrating a comprehensive means of drug molecular design and optimization Four target screening models have been established relying on the company to establish a sound AI new drug screening platform Wisdom pharmaceutical has developed four core pipelines: phosphatase inhibitor screening model, kinase inhibitor screening model, antiviral drug screening model and antibacterial drug screening model At present, phosphatase inhibitor R & D pipeline is the fastest-growing product pipeline of smart Pharma, which has obtained a batch of seedling compounds and is now undergoing optimization and transformation The product can play an important role in many forms of cancer and regulating the immune system, so as to achieve the treatment of solid tumors In addition, there is no relevant inhibitor with the same target on the market Another anti-cancer drug R & D pipeline is a kinase inhibitor screening model This product can inhibit the proliferation and differentiation of tumor cells or promote the apoptosis of tumor cells by inhibiting the activation of related pathways The main indication is hematoma At present, there are many related inhibitors on the market, but drug resistance is common Smart medicine has obtained a number of promising compounds In addition to the two core pipelines in the field of anticancer drugs, smart Pharma is also involved in the research and development of antiviral drugs and anti bacterial drugs The two drug screening models have been developed, and many batches of seedling compounds are being further tested Through smart pharmaceutical's drug screening R & D platform, it can shorten the time needed for drug screening from 2 to 3 years to 2 to 3 months, save the R & D cost of at least two orders of magnitude, and greatly improve the screening accuracy At present, the company has established cooperative research partnership with Melbourne University, Leicester University and Sun Yat sen University in Australia, and is negotiating cooperative development partnership with domestic and foreign pharmaceutical companies (such as Boji pharmaceutical, Takeda, etc.) In terms of business model, smart Pharma will first apply for patent protection for the selected compounds with preliminary inhibition effect, and then it can transfer or jointly develop the compounds The second business model is that the company provides screening services according to the needs of pharmaceutical companies who want to develop a target drug, and the pharmaceutical companies are only responsible for the costs About the future, CEO of smart Pharma said that the company will focus on phosphatase target drugs, patent the selected compounds and actively seek for transfer and development In addition, smart Pharma will also expand its global business and seek multiple partners at home and abroad It is reported that smart Pharma has completed one million level seed round financing and is currently seeking pre-A round financing Jeni turtle