Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstalliTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
The AI did not test 107 million molecules in a laboratory. Researchers used a machine-learning model to predict antibacterial activity across more than 107 million molecules curated in the ZINC15 database, then experimentally tested 23 selected predictions. Eight of those tested compounds showed antibacterial activity; two had potent broad-spectrum activity in the reported experiments.
What did the AI screen?
The study used a machine-learning model to prioritize molecules from ZINC15, a database containing more than 107 million curated molecules. The model predicted which candidates might have antibacterial properties. The scale refers to this computational search, not to laboratory experiments on every molecule. The study abstract describes the search and its results.
Were all 107 million molecules tested?
No. Researchers experimentally tested a discrete set of 23 predictions from the ZINC15 search. Eight of those 23 tested compounds were identified as antibacterial. The paper described those eight as structurally distant from known antibiotics. The result demonstrates how a model can narrow a very large collection to candidates for laboratory follow-up; it does not mean that 107 million compounds were tested, or that the eight candidates are approved medicines.
What did the two strongest ZINC15 candidates do?
Two of the eight compounds showed potent broad-spectrum activity in the reported tests and overcame multiple resistance determinants in Escherichia coli (E. coli). This is an experimental finding about bacterial activity, not evidence that either compound treats infections in people. The study did not establish clinical benefit or regulatory approval. The published abstract reports the activity against resistance determinants.
#1 Best Overall
How does halicin fit in?
Halicin was a separate finding from the same study, identified by searching the Drug Repurposing Hub rather than the ZINC15 collection. It should not be counted among the eight antibacterial compounds identified from the tested ZINC15 predictions.
The researchers reported that halicin was active against a range of pathogens, including Mycobacterium tuberculosis and carbapenem-resistant Enterobacteriaceae. They also reported treatment effects in mouse models of Clostridioides difficile infection and pan-resistant Acinetobacter baumannii infection. These are preclinical results: they do not show that halicin is an effective or approved treatment for people. The study record summarizes these findings.
Rank #2
How the two result groups differ
| Result group | Source | Reported evidence | Experimental stage |
|---|---|---|---|
| Halicin | Drug Repurposing Hub | Activity against a range of pathogens, plus treatment effects in mouse models of two infections | Laboratory pathogen testing and preclinical mouse models |
| Eight antibacterial compounds | 23 experimentally tested predictions from the ZINC15 search | Eight compounds showed antibacterial activity; two showed potent broad-spectrum activity and overcame multiple resistance determinants in E. coli | Experimental testing of selected predictions |
A later paper retrospectively discusses work involving halicin, abaucin from the Drug Repurposing Hub, and antibacterial compounds from the ZINC15 library. That broader summary is separate from the 2020 paper’s specific result of eight antibacterial compounds among 23 tested ZINC15 predictions. The later paper’s abstract provides that context.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
What the study establishes—and what it does not
Published by Jonathan Stokes and colleagues in Cell on February 20, 2020, the study is an example of machine learning supporting early antibiotic discovery by prioritizing candidates for experimental testing. Its figures describe the researchers’ reported results: a computational search across more than 107 million ZINC15 molecules, 23 experimentally tested predictions, and eight antibacterial compounds among those tested.
Rank #3
The reported antibacterial activity and mouse-model findings do not establish that any candidate works in people, is approved, or is currently available as a medicine. The study record does not establish the present clinical-development status of each compound. The Broad Institute lists the paper in Cell, volume 180, issue 4, pages 688–702.e13, DOI 10.1016/j.cell.2020.01.021. See the Broad Institute publication record.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
