GB Sciences developed an AI platform and database to analyze healing properties of plants from nine traditional medicines practices.
GB Sciences’ approach for drug discovery combines artificial intelligence with the power of plants to discover new treatments based upon traditional medicine from around world. Andrea Small-Howard, the company’s chief science officer, and president, stated that they are identifying compounds in plants, and how these treatments have been used in the past. She said that drug discovery was done before computers by using simplified systems.
She explained that traditional drug discovery is focused on finding a single active ingredient to treat one symptom. However, many diseases are complex. The Western approach to illness is to treat each individual symptom. This leads to multiple prescriptions that include medications that are no more needed and side effects.
She said that GB Sciences is searching for multi-component drugs based upon plants that address a realistic view of the body’s workings. “We are trying to address all aspects of an illness simultaneously.
GB Sciences uses an AI approach that involves humans in the loop. Subject matter experts help train the algorithms. Ethnobotanists add information about plants and how they are used to the database. Ethnobotany studies how a society uses local plants for healing illness and injuries.
SEE: Since the pandemic, AI has been adopted by healthcare much faster than ever before.(TechRepublic).
Small-Howard explained that the company is currently collecting data from nine established traditional medicine systems like Traditional Chinese Medicine and Ayurveda. The most difficult part of the work was to create a database that includes all the different types of plants and how they are used.
“We had the data to upload and then we had to make sure that all traditional medicines systems were coded in exactly the same way,” she explained.
Small-Howard explained that the database must also support multiple queries, such a search to find a plant or pain relief, or a search to find a plant in a particular region.
She stated that the database construction was the most unique coding she had to do.
GB Sciences announced that it had filed a provisional application for a patent to protect its proprietary drug discovery platform. This includes a data analysis pipeline and machine learning algorithms. It is designed to identify new active components in traditional, plant-based medicines. The company stated that its goal is to identify complex combinations of active pharmaceutical ingredients derived primarily from traditional plant-based treatments.
Small-Howard stated that the goal of this study is to gather enough evidence to convince Western doctors to prescribe traditional medicines.
She said that areas where these systems are uniquely evolved can be used to predict efficacy.
The company plans on creating a synthetic version, rather than using organic material, of plant compounds. The company is currently researching kava kava (a plant that can be used to treat anxiety) on the islands of the Pacific Ocean.
Small-Howard claimed that the company has employed people from various sources, including NASA.
“We are looking people who are happy adapting what they’ve learned to our work, basically bright people willing take a leap onto something new,” she stated.
The company began with a focus on cannabis. They have developed many treatments using the plant for Parkinson’s disease, chronic pain, cancer and other conditions. Small-Howard stated that the company sold its cannabis assets and invested the profits into the drug discovery and validation pipeline.
GB Sciences was awarded three patents in 2020 for treatments for Parkinson’s Disease, pain, and the anti-inflammatory condition Mast cell Associated Syndrome. GBS also has three corresponding international patents as well as two additional U.S. patents.
A research paper on artificial intelligence in drug development and discovery suggests that AI could reduce human workload and speed up the process. AI is used by traditional pharmaceutical firms to analyze data sets that include millions of compounds.
Another study found that 70% of pharmaceutical companies use AI at least in part. Recruitment and selection of participants in clinical trials is the most common task. This new way of working presents many problems for pharmaceutical companies, such as poor data quality and the need for customizing existing machine learning tools.