2. Past and Present
• In the past few decades
• Few new antibiotics have been developed
• The antibiotics that have been developed are not novel
• Rising issue of antibiotic resistance
• Current methods for screening for new antibiotics are inadequate
3. In Silico Screening
• Machine-learning algorithms
• Screens more than one hundred million chemical compounds to identify
antibiotics that kill bacteria using different mechanisms than those of current
drugs
• Goal:To use artificial intelligence to design new drugs
4. MIT’s Computer Model
• Computer model: Correlated molecular structures of compounds with
characteristics like ability to kill bacteria
• MIT’s Model
• Designed to look for chemical features that make molecules effective at killing
Escherichia coli
• Tested on a large library of compounds
• Identified halicin
5. Halicin
• Chemical structure different from any existing antibiotics
• Strong antibacterial activity
• Low toxicity to human cells
• Killed antibiotic-resistant bacteria:
• Clostridium difficile
• Acinetobacter baumannii
• Mycobacterium tuberculosis
• Disrupts bacteria’s ability to maintain electrochemical gradient across membranes
• ATP cannot be produced
6. Halicin
• Mechanism of chemical makes it difficult for bacteria to develop halicin resistance
Halicin prevented development of
antibiotic resistance in E. coli
Ciproflaxin did not prevent
development of antibiotic resistance in
E. coli
7. Bactericidal Mechanism
• Many antibiotics are bacteriostatic
• Slows down microbial reproduction so that
our own immune systems have a chance to
kill them
• Halicin: Bactericidal
• Halicin killed bacteria with initial cell densities
of 106 CFU/ml
• Potency of halicin decreases as initial cell
density increases
Additional Information
8. Halicin’s MIC and Efficacy
• Minimum inhibitory concentration (MIC): Smallest amount of an agent needed to
inhibit growth of a microbe
• MIC of halicin did not change for bacteria displaying resistance to commonly
used antibiotics
• Classification: Broad-spectrum bactericidal antibiotic
• Pseudomonas aeruginosa
• Lung pathogen
• Encapsulated
• Resistant against halicin
Additional Information
9. Bias
• Not extremely skewed
• Facts were presented straightforwardly
• Neutral presentation of information
• But MIT News was presenting research conducted at MIT itself
• Slight bias could have been present
10. References
1. Trafton, Ann. (2020, February 20). Artificial intelligence yields new antibiotic. MIT
News. http://news.mit.edu/2020/artificial-intelligence-identifies-new-antibiotic-
0220
2. Stokes, J. M.,Yang, K., Swanson, K., Jin,W., Cubillos-Ruiz, A., Donghia, N. M.,
MacNair, C. R., French, S., Carfrae, L. A., Bloom-Ackerman, Z.,Tran,V. M.,
Chiappino-Pepe, A., Badran,A. H.,Andrews, I.W., Chory, E. J., Church, G. M.,
Brown, E. D., Jaakkola,T. S., Barzilay, R., & Collins, J. J. (2020). A deep learning
approach to antibiotic discovery. Cell. 180(4), 688-702.
https://doi.org/10.1016/j.cell.2020.01.021