Researchers from the Hebrew University have developed a unique algorithm that predicts already in the planning stages of the material whether it will have a bitter and repulsive taste, based on its chemical structure. The algorithm, created through Machine Learning, will be able to speed up and reduce drug development processes and prevent trials in the pharmaceutical industry for the purpose of evaluating the bitterness of the substances under development
The research effort in drug development is focused on the development of effective and safe drugs, so that they can cure a disease with a minimum of side effects. The use of sophisticated and automated processes has led to the approval of thousands of drugs by the FDA, with 2020 new drugs approved in 53 alone. However, there is a problem that does not receive an optimal response - many drugs have a very bitter taste that makes it difficult to take them orally. Although putting the medicine into capsules usually solves the taste problem, it is a solution that is not always possible since it increases the medicine and may create real difficulty in swallowing.
While many of us manage to swallow the medicine despite its terrible taste or its problematic size, there are populations that have a very difficult time doing so, including toddlers and adults with swallowing difficulties. Studies have shown that when the medicines had a very bitter taste, a decrease in treatment compliance was observed among children, which led to a health hazard due to the failure to complete the drug treatment. For example, a study published not long ago stated that over 90% of pediatricians report that many children are unwilling to accept medication due to their taste, and this exposes them to health vulnerabilities and the lack of proper medical treatment. Moreover, because of the potential risks caused by the off-putting taste of many medicines, the American FDA recently asked to add a taste note to the list of side effects in prescriptions for children's medicines, to warn parents about those problematic tastes.
Today, the pharmaceutical companies will recognize that there is a problem of a very bitter taste only in the advanced stages of its development, or in the stage of the clinical trials themselves, when the drug is given to thousands of people who are used as prisoners in order to test its effectiveness. In the event that a particularly unusual taste problem is identified, the companies will be forced to go back and change the formulation of the drugs in order to try and mask the terrible taste (which is not always possible), which will result in another delay in the release of the drug to the market, financial losses of millions of dollars and additional and unwanted experiments on animals. For the most part, the pharmaceutical companies would prefer to put a drug on the market even if they already realized that it is very bitter, in the hope that we will deal with the unbearable taste.
In the study group of Prof. Masha Niv, taste researcher at the Institute of Biochemistry, Food and Nutrition at the Faculty of Agriculture, Food and Environment of the Hebrew University, They understood the seriousness of the problem, and succeeded in developing an algorithm that predicts the strong bitterness of a molecule based only on its theoretical chemical structure, even before it was created in the laboratory. The study was recently published in the journal Computational and Structural Biotechnology Journal.
Eitan Margolis, PhD student in Prof. Niv's lab, Tells about the research process and the process of creating the algorithm using a machine learning method regarding intense bitterness: "We collected molecules from theBitterDB established in our laboratory and containing information on over 1000 bitter compounds. We incorporated information about the taste of molecules that we received from our colleagues at the pharmaceutical companyGSK and in the Teva Materials company AnalytiCon Discovery. We taught the computer to understand which combinations of chemical properties are most important in order for a certain substance to be perceived as having a very bitter taste and thus 'trained' it to quickly recognize whether a certain molecule could be very bitter or not. That's basically how we created the algorithmBitterIntense . After testing the model's performance, we saw that it is correct in more than 80% of the cases, which strengthens the potential of its use for the development of taste-friendly drugs - without the need for prior synthesis of the material and its testing in humans or animals. In addition, since it is a computational model, we will be able to continue to improve its accuracy as more information about bitter molecules accumulates."
Contrary to previous scientific beliefs, it was found from the results of the predictions that very bitter drugs do not tend to be more toxic to the liver than less bitter drugs. However very bitter substances appeared to be more toxic to the heart due to their ability to block potassium channels. "The fact that very bitter substances have a higher potential to be toxic to the heart is particularly interesting because bitter taste receptors are also expressed in the heart. Current studies are focused on the physiological functions of taste receptors expressed outside the mouth", explains Prof. Niv.
And what about medicines for the corona disease? Doctoral student Eitan Margolis adds and explains that "we have seen that the prevalence of very expensive drugs among those that are currently in the development stages or approved for the treatment of Corona, is higher than in the general drug group. This result is interesting because the loss of the sense of taste is a very significant and common symptom among those infected with the virus, as studies in Prof. Niv's laboratory have also shown. However, the fact that very bitter medicines can have a beneficial physiological effect, emphasizes the need not to dismiss those very bitter medicines out of hand just based on their taste, but only to prepare accordingly".
In conclusion, the researchers believe that BitterIntense is a method that may reduce financial costs, animal experiments and shorten the time the drug arrives at the clinic. The ability to detect high bitterness in an integrative way during the discovery and development process will help in the development of appropriate drugs for children and geriatric patients. In addition, BitterIntense is also relevant to biotechnology companies and companies that work on the development of new sweeteners (which are sometimes bitter) or natural ingredients that aim to be integrated into food products.
Eitan Margolis shares a personal experience in light of the research: "After the publication of the scientific article, many young people and adults contacted me through social networks, telling me that they very much sympathize with the problem of the bitterness of the medicines. Some of them said that until now they have to chew medicines or eat them with food that has a dominant taste and even then it doesn't always work. Others shared the frustration of giving bitter pills to children who in many cases spit out the pill, and then don't know whether to give the medicine again or in what form and dosage. We hope that thanks to our research, a better answer will be given to the problem of the taste of medicines, and innovative food ingredients."
More of the topic in Hayadan:
