Big data mining thinks in virtual pharmaceutical research

Thinking of big data mining in virtual pharmaceutical research
1. Virtual pharmaceutical research cases based on big data mining
It data mining developed to this day, according to the concept of the current "big" data mining era. Let's start with several related cases.
1.1 Virtual Clinical Test-Big Data Collection
I first look at such a case. In June 2011, Pfizer Pharmaceutical Co., Ltd. announced a "virtual" clinical research, which was a pilot project approved by the US Food and Drug Administration. The first letter was excluded as "Remote". The "Remote" project is the first patient carried out in the United States that only needs to use mobile phones and the Internet instead of repeatedly running the hospital's clinical research. The goal of the project is to determine whether such "virtual" clinical research can produce and traditional clinical research. The same result. Traditional clinical studies require patients to live near the hospital and regularly go to the hospital or clinic for initial examination and multiple follow -up examinations. If this project is effective, it may mean that patients in the United States can participate in many medical research in the future. In this way, the groups that have not been fully represented in the original scientific research projects will be participated, the data collection speed will be greatly accelerated, and the cost is likely to decrease significantly, and the chance of participants withdrawing will likely decrease a lot.
In from the above example, we can see that using the Internet can collect clinical data from super large amounts of patients who are far greater than the number of traditional clinical scientific research samples, and some of these clinical data may come from more convenient wearable health monitoring equipment Essence If such a study, under the circumstances of rigorous scientific and design, effective implementation of quality standards, and effective control of various errors, the efficiency and credibility of scientific research can be significantly improved. As Pfizer's chief medical officer Freida Lewis Holm said: "Let more diverse people participate in research may promote medical progress and bring better results to more patients."
1.2 Virtual drug clinical trial-big data mining
Let's look at another case. In 1992, antidepressant drug Paxil was approved; in 1996, Pravachol, a cholesterol drug, was officially launched. Research from the two drug manufacturers proves that each drug is effective and safe when taking it alone. However, if patients are safe at the same time, no one knows, and few people have thought about it. Researchers at Stanford University in the United States applied data mining technology to analyze the electronic medical records of tens of thousands of patients, and soon discovered an unexpected answer: the blood sugar content of patients taking two drugs at the same time was higher. This has a great impact on patients with diabetes, and excessive blood sugar is a serious health threat to them! Scientists also find hidden laws by analyzing the results of blood glucose testing and prescriptions.
For a single doctor, the patients who have taken these two drugs at the same time are very limited. Although there may be a few diabetic patients that have inexplicable blood glucose, it is difficult for the doctor to realize that this is due to this due to this due to it. Patients were caused by Paxil and Pravachol at the same time. Because this is an implicit law hidden in big data. If it is not for someone to study the safety of the combination of PAXIL and Pravachol, it is difficult for individual doctors to reveal this law. However, there are thousands of clinical drugs. How can we study the safety and effectiveness of the combined application of two or three drugs in any combination? Data mining is likely to be an effective, fast, and active method of exploring a combination of combined application of multiple drugs!
The researchers do not have to call patients to do clinical trials, and it costs too much to do that. The popularity of electronic medical records and computer applications provides new opportunities for medical data mining. Scientists are no longer limited to conducting traditional project research by convening volunteers, but more from experiments in real life, such as screening data and conducting virtual research in a large number of clinical cases. The experimental data of the project is stored in the medical records of many hospitals.
Similar to this case, application data technology enables researchers to find problems that cannot be foreseen when the drug approval is listed. For example, how can a drug affect a specific group of people. In addition, the data mining of medical records will not only bring benefits to research, but also improve the efficiency of the medical service system.
1.3 Virtual drug target discovery-knowledge discovery
Let's look at such a class of research. Generally, the process of new drugs is relatively long, the investment is huge, and the risks are very high. Data show that the average time for new drugs has been 15 years, with an average cost of more than 800 million US dollars. However, due to the poor efficacy of the drug and the high side effects of toxicity, the research and development of many drugs often fails in the clinical stage, causing huge economic losses. As the source of drug research and development, the discovery and identification of drug targets have a significant role in the success rate of drug research and development. With the continuous development of biological information technology, and the increasing growth of protein -based data and chemical genome data, the application data mining technology combines traditional biological experiment technology, which can provide new technical means for the discovery of new targets of drugs, and for target recognition prediction predictions Provide new methods. Building a drug target database, using intelligent computing technology and data mining technology to conduct in -depth exploration of existing drug target data, in order to discover that new drug targets are such a type of research. We also call the knowledge of drug targets.
The discovery of traditional drug targets is usually achieved through a lot of, repeated biochemical experiments, not only high cost, low efficiency, but also very low success. Essence And the application data mining of this automatic, active, and efficient exploration technology can carry out virtual drug target discovery, which not only greatly accelerates the process of drug target discovery, but also greatly reduces the number and cost of biochemistry experiments. At the same time, it has also improved. The success rate of traditional biochemical experiments.
2. Data mining in the application of virtual pharmaceutical research
In the era of big data, pharmaceutical research and development faces more challenges and opportunities. New drugs with competitiveness can be used for data mining technology to carry out virtual medical research and drug research. The application of data mining in virtual pharmaceutical research can be summarized as aspects in the following aspects.
2.1 helps pharmaceutical companies to reduce research and development costs to improve research and development efficiency by predicting modeling. The model is based on the dataset before the drug clinical trial stage and the dataset of the early clinical stage, and the clinical results are predicted as timely and timely. Evaluation factors include product safety, effectiveness, potential side effects and overall test results. By predicting modeling, the research and development costs of pharmaceutical product companies can be reduced. After data modeling and analysis of the clinical results of drugs, they can suspend research on sub -superior drugs or stop expensive clinical trials on sub -superior drugs.
2.2 Through excavation of patient data, evaluating whether recruiting patients meet the test conditions, thereby accelerating the clinical trial process and putting forward more effective clinical trial design suggestions. For example: Patient groups are concentrated through clustering methods, looking for the characteristics of age, gender, condition, and laboratory indicators to determine whether the test conditions are met, or the control group can be better set up according to these characteristics.
2.3 Analysis of clinical trial data and patient records can determine more indications and discovery side effects of drugs. After analyzing clinical trial data and patient records, drugs can be re -positioned or marketing for other indications. Testing data through correlation analysis and other methods may not have thought of some results in advance, and greatly improve the degree of data utilization.
2.4 Collecting adverse reaction reports in real -time or near -time can promote drug warning. Drug alert is a safety guarantee system for listed drugs, monitoring, evaluating and preventing the adverse reactions of the drug. By analyzing the adverse reactions of drugs through big data mining methods such as clustering and associations, the manifestations of medication, disease, and adverse reactions are related to a certain chemical component. For example, cluster analysis of adverse reactions symptoms, correlation analysis of chemical composition and adverse reaction symptoms, etc. In addition, in some cases, clinical experiments have hinted that some situations but do not have enough statistics to prove that evidence can be given based on analysis of clinical trial big data.
2.5 Targeted drug research and development: Personalized drugs through analysis of large data sets (such as genomic data). This application examines genetic variation, susceptibility to specific diseases, and the relationship between special drugs, and then consider personal genetic mutation factors during the process of drug development and medication. In many cases, the patient uses the same medication solution, but the effect is different, partly because of genetic mutation. For different patients of the same disease, they develop different medications or give different usage.
2.6 The combination of pharmaceutical chemical ingredients and pharmacology is excavated to stimulate the inspiration of R