making headlines | Powerful antibiotic discovered by artificial intelligence

progress in: bio-medicine
March 10, 2020




image | above

Pictured is a bio-medical test called an antibiogram. Bacteria is grown in a glass lab tray called a petri dish. A range of antibiotics are applied to the dish — each one is labeled with a button.

After some time, researchers can examine the space around each button. They can see which antibiotics destroyed the bacteria — and which did not.



— note —

Dear readers,

Important breakthroughs happen ever-faster each year. We collect good writing from magazines + newspapers on lead headlines — so you can connect with essential stories on progress.

posted below:

  • a story on progress that’s making headlines
  • facts summary
  • selected reading on this story

topic progress in bio-medicine
story Medical artificial intelligence software discovers powerful antibiotic.

— facts —-

a powerful antibiotic was discovered using an artificial intelligence software tool:

part 1:

  • researchers at MIT used intelligent software to search a library of substances for possible new antibiotics
  • the program successfully identified a compound that has enormous promise for use as an antibiotic
  • the antibiotic they discovered — and named “halicin” — can destroy some of the most dangerous bacteria
  • in lab tests, the drug destroyed several of the world’s most problematic disease-causing bacteria
  • including some strains that are resistant to all known antibiotics
  • it also cleared infections — in 2 lab mouse tests

part 2:

  • the team used a software program designed for “deep learning” — a type of artificial intelligence
  • it can see patterns in vast amounts of data — and select best-fit items that meet a researcher’s criteria
  • the researchers showed the software how to identify types of molecules that destroy bacteria
  • they input — into the program — a library with the molecular features of 2,500 drugs + natural compounds
  • along with data on how well (or not) the substance blocked the growth of the bacteria named escherichia coli
  • so the software program learned what molecular features created good antibiotics
  • next — the team input the program with a library of 6,000 compounds
  • these new substances were under investigation for treating various human diseases
  • the program was set to look for compounds that had anti-microbial features
  • but were also substantially different from current antibiotics
  • in a few hours — the program assessed the compounds + selected some that might be good antibiotics

part 3:

  • the researchers used those selections to test on a range of infections
  • the compound they named halicin — was especially effective
  • they treated numerous drug-resistant infections with halicin
  • halicin killed: mycobacterium tuberculosis — the bacteria that causes the serious lung illness tuberculosis
  • halicin killed: strains of enterobacteriaceae — these bacteria are resistant to carbapenems
  • carbapenems — are a type of antibiotic that are the last resort for enterobacteriaceae infections
  • carbapenems — are commonly used to treat severe + high-risk bacterial infections
  • carbapenems — are reserved for multi-drug-resistant (MDR) bacterial infections
  • halicin cleared: the bacteria clostridiodes difficile + the bacteria acinetobacter baumannii —  in mice
  • halicin could be a much-needed, life-saving antibiotic
  • halicin may be able to treat the worst infections
  • the research team is optimistic

part 4:

  • the World Health Organization said drug-resistant infection is one of the biggest threats to global health
  • using artificial intelligence tech for health-care research is just beginning
  • but major breakthroughs are already being made


image | above

A petri dish in a bio-medical lab holds growing bacteria that was treated with a test of antibiotics.



image | above

The team tested the antibiotic they discovered — and named “halicin” — on infected lab mice, to see if it could clear difficult bacteria from their bodies. The experiment was successful.



selected reading:


1. |

group: MIT
story title: Artificial intelligence yields new antibiotic
read | story

— summary —

A deep-learning model identifies a powerful new drug that can kill many species of antibiotic-resistant bacteria.

motto: mind + hand



2. |

broadcast: BBC
story title: Scientists discover powerful antibiotic using AI
read | story

tag line: To inform, educate, and entertain.


3. |

publication: the Guardian
story title: Powerful antibiotic discovered using machine learning for the first time
read | story

tag line: The world’s leading liberal voice.


4.|

publication: the Conversation
story title: Deep learning AI discovers surprising new antibiotics
read | story

tag line: Academic rigor, journalistic flair.


5. |

publication: Nature • by Springer
story title: Powerful antibiotics discovered using AI
read | story

— summary —

Machine learning spots molecules that work even against “un-treatable” strains of bacteria.


6. |

publication: Financial Times • by Nikkei
story title: AI discovers antibiotics to treat drug-resistant diseases
read | story

— summary —

Machine learning uncovers potent new drug able to kill 35 powerful bacteria.


7. |

publication: Stat
story title: Aided by machine learning, scientists find a novel antibiotic able to kill super-bugs in mice
read | story

tag line: Reporting from the frontiers of health + medicine.


8. |

publication: Singularity Hub • by Singularity Univ.
story title: AI just discovered a new antibiotic to kill the world’s nastiest bacteria
read | story


on the web | pages

IBM | home

IBM | Watson for Health: home
IBM | Watson for Health: learn about AI in medicine

— summary —

Researchers are building better software with artificial intelligence for applications in medicine. Gaining insights into diagnostics, health-care processes, treatment options, and patient outcomes — with the support of computers using a technique called machine learning.

tag line: Making progress in health, together.
tag line: Think.



on the web | pages

AI in HealthCare • by TriMed Media | home

tag line: Innovation to transform health-care.



— notes —

AI = artificial intelligence
ML = machine learning

MDR = multi-drug-resistant

MIT = Massachusetts Institute of Technology
IBM = International Business Machines
WHO = World Health Organization


[ story file ]

story title: making headlines | Powerful antibiotic discovered by artificial intelligence
deck: topic: progress in bio-medicine
folder: stories on progress

[ end of file ]