Drug discovery and development together are the complete process of identifying a new drug and bringing it to market. Discovery may involve screening of chemical libraries, identification of the active ingredient from a natural remedy or design resulting from an understanding of the target. It involves a wide range of scientific disciplines, including biology, chemistry and pharmacology. Development includes studies on microorganisms and animals, clinical trials and ultimately regulatory approval.
Despite advances in technology and understanding of biological systems, drug discovery is still a lengthy, "expensive, difficult, and inefficient process" with low rate of new therapeutic discovery. On average, development of a new drug successfully needs to go through the long process screening about 10000-15000 compounds. It takes about 10-15 years and costs about US$1 billion.
With the accumulation of data from multiple fields, we can accelerate time frames for compound discovery and optimization and enable more effective searches of chemical space by artificial intelligence (AI). Computational drug discovery is an emerging field of machine learning. Our team is dedicated to save time, money and cost in the drug discovery using artificial intelligence.
Now we have established multiple web servers to solve the challenges and difficulties in drug discovery. Firstly, we have built web servers for predicting the bioactivity of given compound, such as the side effect, anti-cancer activity, anti-fungi activity, anti-bacteria activity and anti-HIV activity. Secondly, We have built the web servers for target prediction of compounds based on the sequence and structure separately. Furthermore, we development a method to find the optimal personalized drug(s) to treatment cancer based on personalized genetic background.
Multiple human diseases is currently no effective treatment options, by combining the drug experimental data (including clinical, animal and cytology experiments) and Thomson Reuters IntegrityTM disease targets information, we use a variety of methods to predict the effect of a drug on several human diseases.
Many drug targets is currently unclear, meanwhile, some potential targets have not found a valid drug. We used a drug-target interaction prediction approach based on small molecule features and protein sequences to predict new therapeutic drugs for targets and also explore potential drug-target interaction networks.
Many natural medicines have been successfully used for disease treatment and there more natural compounds are yet to be developed. By combining Traditional Chinese Medicine (TCM) Database@Taiwan and other Chinese herbal medicine data, we conduct drug screening for small active molecules in these natural products.
Personalized treatment is the future direction of medical development. Using existing clinical, exprimental and pharmaceutical data, we conducted a personalized therapeutic drug prediction and efficacy evaluation based on each person's physiological indicators, transcriptome profile and mutation information.