Courtesy Photo | Encrypted peptides were identified within protein sequences from the human proteome...... read more read more
Subscribe
Analysis of the human proteome may lead to potential antimicrobial treatments.
Antibiotic-resistant diseases are some of the most significant public health issues today. An artificial intelligence (AI) and machine-learning (ML) tool recently developed at the University of Pennsylvania (UPenn) searches for antibiotics that lie hidden within proteins naturally produced by the human body. These small protein antibiotics called antimicrobial peptides (AMPs) could be the key to developing new drugs that are effective against the increase in antibiotic resistance that threatens the health and readiness of the Joint Force.
Antibiotic resistance results from genetic or environmental adaptations that trigger new or preexisting defense mechanisms in bacteria, allowing them to evade the toxic effects of an antibiotic, persist, and cause disease. Decades of poor antibiotic stewardship and misuse sped up this naturally occurring phenomenon that caused the rise in antibiotic-resistant bacteria. Drug-resistant genes are more prevalent worldwide, which increases the opportunity for bacteria to acquire antibiotic resistance traits faster than ever.
Developing new antibiotics capable of eliminating drug-resistant bacteria is critical to our survival. The existing antibiotic drug development pipeline is expensive, lacks sufficient profit incentives necessary for motivating investments by the pharmaceutical industry, and is time-consuming. Drug development often takes over a decade before receiving Food and Drug Administration approval.
The Defense Threat Reduction Agency’s (DTRA) Chemical and Biological Technologies Department in its role as the Joint Science and Technology Office (JSTO) for the Chemical and Biological Defense Program invested in the UPenn research that led to the discovery of thousands of new human-derived AMPs hidden within sections of known proteins which can be used in new antibiotic development.
Scientists use AI/ML search tools to speed up the drug-discovery process by quickly mining through large amounts of data, which can expedite the development of new broad-spectrum therapeutics. Since there were no tools for rapid identification of human-derived antibiotics, the UPenn team created an ML tool that scanned the sequences of all human proteins, looking for AMPs. The resulting search led to identifying 2,603 new AMPs from which the team selected 55 candidate peptides that were synthesized, characterized, and validated in subsequent laboratory tests. A study of the biochemical characteristics of these 55 peptides demonstrated that they were different from all known AMPs to date and had biological functions unrelated to the immune system. They demonstrated significant clinical relevance as 63% of the synthesized AMPs suppressed growth of eight bacterial strains on the World Health Organization’s watch list.
The UPenn report featured on the cover of the January 2022 issue of Nature Biomedical Engineering demonstrated their ground-breaking approach that uses an AI/ML algorithm to search within the human proteome to discover unique antimicrobials. The article provides the researchers’ ML code that can be used by other scientists to broaden the search for AMPs in other organisms to discover new naturally occurring AMPs for antibiotic development. This could help create a new panel of antibiotics that expands the available antimicrobial spectrum to aid DTRA JSTO’s efforts in protecting the Joint Force from antibiotic-resistant infections.
POC: Dale Taylor, dale.e.taylor4.civ@mail.mil
Drug Development Process
| Date Taken: | 11.10.2022 |
| Date Posted: | 11.10.2022 18:05 |
| Story ID: | 433115 |
| Location: | FT. BELVOIR, VIRGINIA, US |
| Web Views: | 331 |
| Downloads: | 1 |
This work, Natural Proteins. Artificial Intelligence., by Diane Williams, identified by DVIDS, must comply with the restrictions shown on https://www.dvidshub.net/about/copyright.
No keywords found.
