YouTube

Got a YouTube account?

New: enable viewer-created translations and captions on your YouTube channel!

English subtitles

← The medical potential of AI and metabolites

Get Embed Code
35 Languages

Showing Revision 5 created 10/29/2019 by Erin Gregory.

  1. In 2003,
  2. when we sequenced the human genome,
  3. we thought we would have the answer
    to treat many diseases.
  4. But the reality is far from that,
  5. because in addition to our genes,
  6. our environment and lifestyle
    could have a significant role
  7. in developing many major diseases.
  8. One example is fatty liver disease,

  9. which is affecting over 20 percent
    of the population globally,
  10. and it has no treatment
    and leads to liver cancer
  11. or liver failure.
  12. So sequencing DNA alone
    doesn't give us enough information
  13. to find effective therapeutics.
  14. On the bright side, there are
    many other molecules in our body.

  15. In fact, there are
    over 100,000 metabolites.
  16. Metabolites are any molecule
    that is supersmall in their size.
  17. Known examples are glucose,
    fructose, fats, cholesterol --
  18. things we hear all the time.
  19. Metabolites are involved
    in our metabolism.
  20. They are also downstream of DNA,
  21. so they carry information
    from both our genes as well as lifestyle.
  22. Understanding metabolites is essential
    to find treatments for many diseases.
  23. I've always wanted to treat patients.

  24. Despite that, 15 years ago,
    I left medical school,
  25. as I missed mathematics.
  26. Soon after, I found the coolest thing:
  27. I can use mathematics to study medicine.
  28. Since then, I've been developing
    algorithms to analyze biological data.
  29. So, it sounded easy:
  30. let's collect data from all
    the metabolites in our body,
  31. develop mathematical models to describe
    how they are changed in a disease
  32. and intervene in those
    changes to treat them.
  33. Then I realized why no one
    has done this before:

  34. it's extremely difficult.
  35. (Laughter)

  36. There are many metabolites in our body.

  37. Each one is different from the other one.
  38. For some metabolites,
    we can measure their molecular mass
  39. using mass spectrometry instruments.
  40. But because there could be, like,
    10 molecules with the exact same mass,
  41. we don't know exactly what they are,
  42. and if you want to clearly
    identify all of them,
  43. you have to do more experiments,
    which could take decades
  44. and billions of dollars.
  45. So we developed an artificial
    intelligence, or AI, platform, to do that.

  46. We leveraged the growth of biological data
  47. and built a database of any existing
    information about metabolites
  48. and their interactions
    with other molecules.
  49. We combined all this data
    as a meganetwork.
  50. Then, from tissues or blood of patients,
  51. we measure masses of metabolites
  52. and find the masses
    that are changed in a disease.
  53. But, as I mentioned earlier,
    we don't know exactly what they are.
  54. A molecular mass of 180 could be
    either the glucose, galactose or fructose.
  55. They all have the exact same mass
  56. but different functions in our body.
  57. Our AI algorithm considered
    all these ambiguities.
  58. It then mined that meganetwork
  59. to find how those metabolic masses
    are connected to each other
  60. that result in disease.
  61. And because of the way they are connected,
  62. then we are able to infer
    what each metabolite mass is,
  63. like that 180 could be glucose here,
  64. and, more importantly, to discover
  65. how changes in glucose
    and other metabolites
  66. lead to a disease.
  67. This novel understanding
    of disease mechanisms
  68. then enable us to discover
    effective therapeutics to target that.
  69. So we formed a start-up company
    to bring this technology to the market

  70. and impact people's lives.
  71. Now my team and I at ReviveMed
    are working to discover
  72. therapeutics for major diseases
    that metabolites are key drivers for,
  73. like fatty liver disease,
  74. because it is caused
    by accumulation of fats,
  75. which are types
    of metabolites in the liver.
  76. As I mentioned earlier,
    it's a huge epidemic with no treatment.
  77. And fatty liver disease
    is just one example.

  78. Moving forward, we are going to tackle
    hundreds of other diseases
  79. with no treatment.
  80. And by collecting more and more
    data about metabolites
  81. and understanding
    how changes in metabolites
  82. leads to developing diseases,
  83. our algorithms will get
    smarter and smarter
  84. to discover the right therapeutics
    for the right patients.
  85. And we will get closer to reach our vision
  86. of saving lives with every line of code.
  87. Thank you.

  88. (Applause)