On September 26, 2020, XtalPi, an artificial intelligence drug research and development company driven by digitization and intelligence, announced that it has over-completed a $318.8 million Series C financing, creating the world’s highest financing record in the field of AI drug research and development.
According to XtalPi, Rongde funds will be used to further develop the smart drug R&D system, build a new AI-enabled digital drug R&D infrastructure from the three dimensions of computing power, algorithm, and data, and serve global pharmaceutical companies, biotechnology companies and cooperation Partners to meet the urgent needs of the pharmaceutical industry to increase efficiency and speed.
After the huge C round of financing, this company from the Massachusetts Institute of Technology has been promoted to the first echelon industry of AI new drug research and development, becoming one of the most influential companies in this field on a global scale. After joining a number of well-known investment institutions at home and abroad, XtalPi is expected to significantly strengthen its R&D capabilities and achieve business collaboration with more major international pharmaceutical companies in drug R&D.
Digging deep into the three elements of AI, XtalPi has already built up barriers
Algorithms, data, and computing power are the three core elements that affect the development of artificial intelligence technology. XtalPi can stand out from many AI pharmaceutical companies because it has worked hard in these three elements.
In terms of algorithms, XtalPi has completed a reserve of hundreds of algorithms based on dual cores of physical theory and artificial intelligence theory. Its physics-based theoretical framework covers algorithms from quantum mechanics to classical mechanics. Through the construction of multi-scale and multi-dimensional models, it can accurately describe the interaction between drug molecules and human proteins from the bottom; its artificial intelligence theoretical framework is completed The AI reinforcement and cross-validation model construction between different feature vectors corresponding to various key properties of drugs is presented.
In the next step, XtalPi will also build a digital twin R&D system that closely integrates modern laboratories and digital R&D platforms to realize efficient interactive iteration of experimental exploration and drug simulation algorithms.
In terms of data, XtalPi uses the data lake as the main method of data governance, so that the huge amount of data generated by calculation is not only directly presented as the calculation result, but also applied to scenarios such as data analysis and machine learning. XtalPi also uses a special data strategy to accumulate a large amount of underlying data in the chemical space of drug molecules through quantum mechanical calculations.
Currently, XtalPi’s data accumulation is close to PB. With the development and deployment of the digital twin R&D system and the continuous increase in the amount of calculations, XtalPi will support the continuous innovation and upgrade of its algorithms with rich drug R&D real-world data, challenge more R&D bottlenecks, and solve urgent research problems in the drug R&D industry.
In terms of computing power, XtalPi is based on a multi-cloud architecture model and has built a computing platform that can dispatch massive resources globally on AWS, Tencent Cloud, Google Cloud and other public clouds. Using cloud computing elastic expansion and parallel computing capabilities, the platform can create a super cluster of nearly one million computing cores within a few hours. This platform allows XtalPi to simultaneously carry out ultra-large-scale drug molecular discovery and screening projects, and supports the large-scale application of its drug AI research and development technology.
Drug R&D “new infrastructure” kicked off, XtalPi uses AI to promote the pharmaceutical industry revolution
The outbreak of the new crown epidemic in 2020 will undoubtedly have a huge impact on the living order and economic order of people all over the world. But things always have two sides. While causing damage, the new crown epidemic has also sounded the alarm for the construction of a global epidemic defense system, added a catalyst to the development of the pharmaceutical industry, and forced companies to accelerate research on a more efficient drug R&D system.
Artificial intelligence is undoubtedly one of the important technologies to break the existing drug research and development system. With the support of high computing power, this technology can accelerate the drug development steps that rely on trial and error in the new drug development process, and use accurate and fast algorithms to predict the results. , To help pharmaceutical companies improve R&D efficiency, success rate, reduce risks, and ultimately accelerate drug R&D.
During the epidemic, XtalPi found that only by combining the AI model + physical thinking in depth, can it meet the requirements for speed and accuracy of drug development. Here, XtalPi uses cloud supercomputing to support physical and chemical algorithms to build a virus model, and then starts from the structure, through AI to accelerate the search for FDA-approved drugs with antiviral activity, and provide the most direct and effective drug screening strategy during the epidemic.
On January 20, NCBI announced the gene sequence of the new coronavirus for the first time, which provided a basis for constructing a protein structure model of the virus. Drug screening and design based on structure provide a method with higher accuracy and higher success rate for calculation-based screening methods. On the same day, XtalPi established a new coronavirus research team. Based on the gene sequence and the homologous protein crystal structure of the SARS virus, it built a three-dimensional structure model of the key protein on the new coronavirus SARS-CoV-2, including those that affect the virus replication process. 3CLpro and PLpro, and Spike protein receptor binding region closely related to infected patients.
On February 2, XtalPi released these high-precision models, a large amount of calculation data, and research results on the self-media and company websites, and they were downloaded thousands of times within a few days. These models provide a basis for studying the molecular mechanism, design and screening of viruses. As the research progresses, XtalPi is still updating shared content that can be downloaded by peers from time to time.
After the completion of this round of financing, XtalPi will continue to contribute valuable research results to the industry, and fully deploy AI-enabled new digital drug R&D infrastructure, promote the transformation of the pharmaceutical industry to a smarter drug R&D model, and help more pharmaceutical companies, Biotechnology companies use artificial intelligence to develop more efficient drugs that benefit patients around the world, and also contribute to the digital upgrade of artificial intelligence in China’s pharmaceutical industry.
“In the past few years, we have focused on building and improving the self-developed ID4 smart drug R&D platform. Through cooperation with major pharmaceutical companies and biotech companies around the world, we have continuously verified the advantages of the ID4 platform in terms of cost, speed and success rate. And great potential, and achieved many milestone progress.” XtalPi co-founder and chairman Wen Shuhao introduced. “It’s an honor that as an advocate and pathfinder in the field of AI drug research and development, we have been recognized by more and more investment institutions in terms of technology, team and social value.”
At the same time, Wen Shuhao also said: “‘Patients have their medicine’, five simple words, carrying the vision of XtalPi, which makes us the simplest wish: We hope that all human individuals suffering from illness, regardless of age, nationality or skin color The rich and the poor can be treated equally if there are diseases, and good medicines are available and affordable.”
“However, this is not an easy task. The movie “I am not the God of Medicine” resonates widely and reflects this reality. Among the leading companies in the pharmaceutical industry, the rate of return on new drug R&D since 2015 has dropped. To less than 5%, the failure rate and cost of drug development are getting higher and higher, making good drugs more and more expensive; in addition, there are still a large number of diseases without drugs that can be cured.
Artificial intelligence is completely possible to change this status quo. XtalPi hopes to use AI as a technology with temperature to improve the efficiency and success rate of drug research and development through algorithms, computing power, and data, reduce drug research and development costs, and explore a broader chemical space. In the end, it will benefit patients all over the world, and bring more specific medicines to more people in need.