AI compound screen finds new antibiotics against drug-resistant gonorrhea
New CapabilitiesA machine-learning model scanned millions of molecules and surfaced two that kill a pathogen running out of treatments
June 29th, 2026: MP20 and A1 detailed in journal and Nature coverageNew here? Follow stories to track developments over time. Create a free account to get updates when stories you care about change.
Overview
Updated Jun 30Gonorrhea has shrugged off nearly every drug thrown at it. Now a machine-learning model has scanned about 6 million chemical structures and pulled out two compounds that kill the bacterium in ways existing antibiotics do not.
The work, led by James Collins at Harvard's Wyss Institute with MIT and the Broad Institute, was published in Science Translational Medicine. The lead compound killed the bacterium in a chip lined with living human cells. Both compounds are years from a pharmacy shelf, but they give doctors a new line of attack on a pathogen the World Health Organization lists as a high-priority resistance threat.
Why it matters
Gonorrhea beats every drug class eventually; the AI compounds add two new mechanisms before the current options start failing.
Questions about this story
Free account needed to ask — your question is kept and asked for you right after sign-up. Answers are public.
No questions yet — be the first to ask.
Key Indicators
Voices
Curated perspectives — historical figures and your fellow readers.
Play
Exploring all sides of a story is often best achieved with Play.
Higher or Lower
A number from this story, against one from elsewhere in the news — guess which is bigger, then keep the chain going. 5 rounds, 3 strikes; a miss costs a strike and resets your streak.
Keyboard: ↓/L lower · ↑/H higher
0 points — sign up to put that on the leaderboard.
Timeline
Order five events from this story, oldest at top. Each in the right slot scores 1 — neighbours within one slot count too. Your previous result — green ✓ for exact slots, yellow ~ for off by one. Cards now in true chronological order.
Sign up to save your score and track a streak across stories.
Connections
Sixteen names from the news. Find the four hidden groups of four. Four mistakes max.
Sign up to keep a daily streak — a new puzzle lands every day.
Exit debate?
Your progress in this debate will be lost.
- 1 Two AI personas square off on this story.
- 2 You predict who'll win each round — correct picks earn XP.
- 3 One crossfire question is yours to fire. Pick it carefully.
Couldn't generate a topic
Select Your Champions
Choose one persona for each side of the debate
DEBATE TOPIC
Choose personas with different perspectives for a more dynamic debate.
Select debater for this side:
No debate personas available right now.
Select debater for this side:
No debate personas available right now.
Who's Got This Round?
Make your prediction before the referee scores
The referee scores both sides on
Round Results
Set the Crossfire
Pick the question both personas must answer in the final round
Debate Oracle! You called every round!
Sharp Instincts! You know your debaters!
The Coin Flip Strategist! Perfectly balanced!
The Contrarian! Bold predictions!
Inverse Genius! Try betting the opposite next time!
XP Breakdown
Prediction History
People Involved
Organizations Involved
A Harvard-affiliated institute that builds biologically inspired engineering tools, including the organ-on-a-chip used to test the compounds.
The United Nations health agency that ranks drug-resistant pathogens and tracks gonorrhea resistance through its surveillance program.
Timeline
February 2020 June 2026
-
MP20 and A1 detailed in journal and Nature coverage
Latest PublicationNature covers the Science Translational Medicine paper. MP20 disrupts the bacterial membrane and damages DNA; A1 blocks an enzyme needed to build the cell wall.
-
Gonorrhea compounds reported
DiscoveryCoverage details a graph neural network trained on about 38,650 molecules, then used to screen 6 million, yielding 83 active candidates and two leads.
-
FDA approves zoliflodacin and gepotidacin for gonorrhea
RegulatoryThe FDA approved zoliflodacin (Nuzolvence) and gepotidacin (Blujepa) on consecutive days—the first new oral gonorrhea treatments in nearly 40 years. A phase 3 trial showed zoliflodacin's single 3 g oral dose cured 91% of patients and matched the ceftriaxone-plus-azithromycin standard. Gepotidacin cured 93%.
-
Generative models design compounds from scratch
MethodMIT researchers use generative AI to design new antibacterial structures atom by atom, rather than only ranking existing molecules.
-
AI surfaces abaucin
DiscoveryThe lab reports abaucin, a narrow-spectrum compound against Acinetobacter baumannii, found by screening thousands of molecules with a trained model.
-
Deep learning finds halicin
DiscoveryCollins's group reports halicin, a compound found by a neural network, killing several drug-resistant bacteria in mice. It is the method's first proof.
Historical Context
3 moments from history that rhyme with this story — and how they unfolded.
Halicin discovery (2020)
Collins's team trained a neural network to predict antibacterial activity, then ran it across a drug-repurposing library. It flagged halicin, a molecule once studied for diabetes, which killed many drug-resistant strains in mice.
The find drew wide attention as the first antibiotic surfaced by deep learning.
It established the screen-and-confirm method the gonorrhea work now extends to a harder target.
The gonorrhea study uses the same core idea, scaled from a small library to roughly 6 million compounds.
Abaucin against Acinetobacter baumannii (2023)
The lab screened about 7,500 molecules with a trained model and found abaucin, a compound that hits one dangerous hospital pathogen while sparing others. Narrow targeting can slow resistance and protect helpful bacteria.
Abaucin worked against the target in mice and lab tests.
It showed AI screens could deliver precise, narrow-spectrum drugs, not just broad killers.
MP20 and A1 follow the same narrow-targeting logic against a single high-priority pathogen.
Zoliflodacin oral antibiotic trial (2024)
A late-stage trial showed zoliflodacin, a first-in-class oral drug, matched standard injectable treatment for uncomplicated gonorrhea. It was the first genuinely new gonorrhea antibiotic class to reach that stage in decades.
Results supported regulatory review of a new oral option.
It showed new mechanisms can still beat a pathogen that has defeated older drugs.
It marks the slow, expensive path the AI-found compounds must travel to reach patients.
