Fighting Superbugs By Searching the Microbiome for New Antibiotic Treatments

Designed to Cure


More infectious bacteria than ever are acquiring antibiotic resistance, so scientists are searching the microbiome for compounds that have antibacterial properties.  Image Credit: Flickr user Esther Simpson

As more strains of bacteria develop antibiotic resistance, scientists are searching for new sources of antibiotics. While most current antibiotics are derived from compounds produced by soil bacteria, new research indicates that some bacteria in the microbiome also produce compounds with antibiotic properties.1 In fact, a recent article in the journal Nature reported that a strain of bacteria found in the human nose can produce lugdunin, a cyclic peptide that effectively treats infections caused by a strain of Staphylococcus aureus. Moreover, unlike many common antibiotic treatments, the compound is not likely to cause S. aureaus to develop resistance.2

This study offers proof of principle that some bacteria in the human microbiome can be used as sources of antibiotics.3 In order to determine which bacterially produced compounds are the most promising candidates for use in antibiotic drugs, researchers at life sciences firms can analyze their properties using modern modeling software.

Understanding the Properties of Candidate Compounds

The potential for a particular compound to work in an antibiotic drug depends on its chemical properties. With computer modeling, scientists can analyze the relationship between a candidate compound’s molecular structure and its behavior. If the compound is already known to be able to kill bacteria, scientists may be able to use software to screen larger numbers of compounds more quickly. Scientists may also be able to conduct complex computer simulations on compounds derived from bacteria in the microbiome in order to determine how they will behave in certain environments or how they might interact with particular surface markers on antibiotic-resistant, infection-causing bacteria. Simulations studies may also be able to test different delivery methods for new antibiotic drugs in order to predict which will be the most effective.

Another way to predict whether a  new compound can work as an antibiotic drug is by comparing it to compounds used in existing antibiotic treatments. With three-dimensional modeling, scientists can analyze two compounds side by side and try to identify similarities and differences at the atomic level. Not only could this information be used to predict whether a candidate might have antibiotic properties similar to those of a current drug, but it could also offer insight into how quickly the target bacterium might develop resistance, depending on how closely the candidate resembles the current drug.

Collaboration Between Scientists

Because the microbiome has yet to be thoroughly searched for bacteria that produce potential antibiotics, it is essential for scientists to be able to share their findings as they navigate this relatively uncharted territory. That way, they can learn from each other’s discoveries. For instance, when one researcher finds a compound that is predicted to have antibiotic properties, experimental results can be shared with other lab members so that they can compare it to the compounds they are screening and identify new candidates with similar properties.

It can also be helpful for scientists to collaborate with each other on the modeling process itself. After data scientists develop a basic model, they can share it with researchers working on particular microbiotic compounds, who can use the guided model-building tools provided by modern software to tailor the procedures to their specific projects.  By giving end users these tools, the software ensures that they are using the most effective possible method for picking out the most promising antibiotic candidates.

Using Software to Improve Research Efficiency When Searching the Microbiome for Antibiotic Treatments

As more infectious bacterial strains develop resistance to existing antibiotic drugs, researchers who are exploring the microbiome for alternatives must be able to screen compounds, conduct tests on promising candidates and start developing new drugs as quickly as possible. Using software to model potential antibiotics, rather than starting with bench tests, can reduce the number of time-consuming bench tests that need to be conducted later, since simulations alone are often sufficient for highlighting many key chemical properties and behaviors. Not only does this speed the development process for an antibiotic drug, but it also frees lab members to return to the search for new candidate compounds sooner.

Most modern modeling software also allows for the automation of repetitive modeling tasks that previously slowed down simulation setups. Now, scientists can spend their time designing tests and analyzing results rather than tinkering with the mundane technicalities of modeling. Without these unnecessary time lags, the company can bring new antibiotic drugs to market more quickly than ever.

BIOVIA Designed to Cure is a software suite that can support innovative drug research and development. The software’s in silico modeling and advanced analytical capabilities can help researchers at life sciences firms gain insight into drug candidates at the molecular level. Contact us today to find out more about our software offerings.

  1. “A Systematic Analysis of Biosynthetic Gene Clusters in the Human Microbiome Reveals a Common Family of Antibiotics,” September 11, 2014,
  2. “Human commensals producing a novel antibiotic impair pathogen colonization,” July 28, 2016,
  3. “‘Nose-y’ Bacteria Could Yield A New Way To Fight Infection,” July 27, 2016,