#108: Alex Zhavoronkov - Revolutionizing Medicine: How AI Will Cure Aging

Are you ready to dive into the future of medicine and the revolutionary advancements being made with artificial intelligence? In this episode, I sit down with Alex Zhavoronkov, the founder and CEO of Insilico Medicine, a global company focused on the discovery of novel therapeutics utilizing cutting-edge technology in healthcare.

From longevity research to anti-aging treatments, Alex shares his expertise on the
most impactful field of our time, discussing his work in protein target
discovery and the design of small molecule drugs using generative AI.

But it doesn't stop there. Alex also shares insights into the development of aging
clocks, biomarkers of aging, and precision medicine. Discover how generative AI
is transforming the pharmaceutical industry, improving clinical trials, and
providing a revolutionary new approach to drug discovery methods.

Don't miss out on this incredible opportunity to learn from one of the leading
experts in AI and healthcare innovation. Tune in now and join us on this
fascinating journey into the future of medicine.

💡 LINKS TO MORE CONTENT
If you like the episode, become a subscriber and support the show: https://lsg2g.substack.com/subscribe
Watch on Youtube
Alex Zhavoronkov
Christian Soschner
Check out the organizations that make the podcast possible

📖 Memorable quotes:
(10:50) “Insilico Medicine synthesized and tested the first AI generated drug in 2017.”

(23:09) “Generative AI in Drug Discovery and Drug Development has the potential to improve the probability of success by 50%”

(24:00) “Generating 11 preclinical candidates in 2.5 years - is a pretty cool number.”

(29:48) “We developed a clinical candidate in 18 months at a 3 million dollar cost with the help of AI. The traditional route takes 2 years at about 400 million expeonses.”

(01:08:48) “Generative AI will transform our lives beyond recognition in every way.”

⏰  Timestamps:
(02:00) Introduction to Alex Zhavoronkov at World Government Summit 2023 

(03:00) About World Government Summit 2023 

(04:20) Alex's longevity research at Insilico Medicine 

(08:20) Accelerating drug discovery with AI 

(13:00) AI in synthetic biologic data 

(15:00) Generative AI in reducing cost and increasing output 

(23:30) Generative AI in increasing probability of success 

(25:00) Why traditional drug R&D takes over 10 years and 2 billion 2010 dollars 

(28:45) AI-driven drug discovery at Insilico Medicine 

(30:30) Insilico Medicine's pipeline and case study 

(38:09) Regulators' perception of AI in drug development 

(45:26) Generative AI in clinical development

 (49:15) Audience question on AI and microbiome research 

(53:30) Operating a company between China, Middle East, and North America 

(57:54) Insilico Medicine's fully automated robotics lab 

(01:03:15) Ageing as the most important disease in the world (

01:08:00) Future impact of generative AI on medicine

Transcript

what if I told you that there is a way to reverse the aging process and

0:06

potentially extend our lives by decades sounds too good to be true so we made a

0:13

little bit of a history generate the way I is going to transform our life uh beyond recognition in every way I think

0:20

that we can at least improve the probability of success by 50 and that's

0:25

more more important than time and cost so in two years uh well two and a half

0:30

years 11 pre-clinical candidates it's a pretty cool number so that is how long it took us to get to

0:39

pre-clinical candidate stage with our anti-fibrotic uh and again this is the traditional approach this is our

0:46

approach so under 18 months uh the ability to very rapidly predict success

0:53

or failure and also interpret I think that's uh that's key well today's guest Alex savaronkov CEO of in Silicon

1:01

medicine is animation to make this a reality with the help of artificial

1:06

intelligence his team is working to uncover the secrets of aging and create

1:11

drugs that could not only extend our lifespans but also prevent age-related

1:17

diseases like Alzheimer's or cancer in this podcast you learn about the

1:23

incredible potential of artificial intelligence in aging research and how

1:28

in Silicon medicine is leading the way Alex explains the potential to reduce drug development costs increase the

1:35

probability of success of track Discovery and even create drugs for

1:41

diseases we once thought were untreatable get ready to have your mind

1:46

blown by the groundbreaking research and insights shared in this episode don't

1:52

forget to hit the Subscribe button and join us on this journey to revolutionize

1:58

the way we age your life directly from the world's government government

2:04

Summits dedicated rights that's correct so I'm dialing in from the world

2:10

government Summit I apologize for any background noise because I had to step

2:16

out of the meeting I am here it's uh there are no meeting spaces

2:24

um so very happy to connect so my name is Alex chavarenkov I'm the founder and CEO of a company called in silica

2:30

medicine uh which is a global company so that's what that's why I am in the Middle East

2:37

right now or one of our sites is in Abu Dhabi and uh which is a fantastic place

2:42

to set up I and right now I'm presenting from Dubai what is the world government Summit all

2:49

about uh it's all about bringing together uh

2:54

world leaders from all around the world to um ensure that they can achieve common

2:59

goals uh they also invite business people and scientists to ensure that uh

3:06

they have the support from the private and public sector for all kinds of uh discussions

3:15

about the environment Healthcare and other areas where Global collaboration is

3:21

extremely important so it's practically similar to the world economic Forum it's

3:27

uh that's just that's correct it's the Middle East version of that yeah it's the same food it's it's a good thing to

3:34

have that we are part of the summit now with this podcast so thanks for for coming online Alex I read in your uh in

3:40

your bio that stats directly jump right into it and into what you're doing with in silica research uh I found a lot of

3:47

uh articles on the internet where you described that you come from longevity research I think we have that in common

3:54

I did something similar not in the academic field more in the field of martial arts because the most important

3:59

question I had was in that it was back in the 90s how can I wait to get sick

4:04

and since in the 90s in Austria there was not many people really interested in

4:12

this topic I switched model martial arts to find how I can improve my life and now it's called the pink long charity

4:17

research what did you do in this field so I am participating in this field more

4:24

from the academic perspective rather than for from the perspective of

4:29

self-improvement I I think that longevity in general is probably the most impactful field

4:36

anybody can be focused on right because uh if you figure out a way to extend the

4:44

life of our people just by one year for everybody on the planet you generate 8

4:49

billion life here right so that's more than any surgeon or a doctor can achieve

4:56

in their lifetime so I am focused on the discovery of Novel Therapeutics

5:02

uh that may address certain Hallmarks of Aging so we look at both uh um uh we

5:11

look at both uh protein Target Discovery so identifying the uh various possible

5:17

protein switches that might be important in uh aging and Longevity uh and then we

5:24

also have a way to design novel Therapeutics novel small molecule

5:29

drugs that go after those targets uh so we have uh kind of two hands one

5:35

searches for promising areas to intervene them and another hand which

5:40

allows you to very rapidly generate the chemical matter um that can be turned into drug

5:48

uh and for that I study um and I contribute to research in

5:55

artificial intelligence that works both in biology and chemistry

6:00

uh we have a platform called Pharma AI it and silica that helps accelerate uh

6:06

drug Discovery r d but many of the programs that we go after are in age

6:11

Associated diseases and Longevity I we also contributed quite a bit to the

6:17

development of a variety of what is called what is called Aging clocks

6:23

on the biomarkers of Aging utilizing artificial intelligence so you can

6:30

actually train train the neural networks uh um or other AI models uh that allow

6:37

you to predict chronological age in a healthy state of the patient let's call

6:42

it a biological age or you could predict uh um uh subjective age so how old do you feel

6:51

or even some other features like for example perceived age how all do other

6:59

people think you are right if you are talking about Imaging biomarkers or psychological biomarkers so we can look

7:06

at many many different data types to build those comprehensive age predictors uh and later we can design certain

7:14

interventions for uh biomarkers that could yield valuable targets either

7:20

protein targets or even sometimes non biological targets um so looking at both again

7:27

interventions and diagnostic applications that's a great thing

7:34

um I found your company when I was doing some research on LinkedIn about another topic chat GPT and then I came across an

7:41

article that's uh like you know how it's uh like where you basically compare Your

7:47

solution and track development and track Discovery to check GPD and Dali from

7:53

open AI uh and defaults to that in moment was so now that I know how to

7:58

work with chatgpt and it can produce articles and questions and stuff like that uh is it really possible to have an

8:06

artificial intelligence player just enter the questions that I want to cure cancer and yeah artificial intelligence

8:11

then spits out the drug plans the Target and everything is solved what is the state of the art in uh generative

8:18

artificial intelligence in your field so sure well again generative AI as a

8:25

term has been around for quite a while uh I think that the term generative

8:32

artificial networks um are generated the cell network sorry was uh guns I was coined by Ian

8:39

Goodfellow in 2014 uh in a very famous paper that he co-authored with Joshua

8:44

Benjo probably the highest sighted researcher in Ai and uh um in that paper

8:53

they've described how to utilize generator the serial networks to imagine

8:59

handwriting written characters for example since then in just a year the

9:04

technology has progressed to the level where you could generate pictures uh that are of very high quality where we

9:12

can describe what you want to see and Gyan will create it for you uh that was

9:18

the year when we got into uh generated the serial number works as well with molecules so just like with a picture or

9:25

with a handwritten character you can describe the molecule you want and the

9:32

yarn will generate it for you with the desired properties using multi-parameter optimization techniques uh and uh that

9:40

since since 2016 this technology has progressed uh pretty dramatically right

9:46

so in 2020 you could generate pictures of people that are indistinguishable

9:52

from photographs right or almost indistinguishable uh and uh nowadays uh

9:58

with this tools like dally a mid-journey uh generative Technologies uh now also

10:05

Transformer neural networks and other forms of generative AI uh became

10:11

consumerized so now consumers got access to those tools that we were using

10:16

professionally for quite a quite a while uh and they have propagated uh into many

10:23

Industries uh so mid-journey and Dali for example are now used to generate

10:28

very valuable uh images from a single prompt uh chat GPT is now outperforming

10:36

humans in some of the writing tasks especially when language is not your first language uh but in our field in

10:46

drug Discovery this technology has progressed quite a bit so we synthesized

10:53

and tested our first AI generated molecule in 2017.

10:58

that led to some considerable investment but also I established us as a player in

11:04

this field uh so there we designed small molecules against a specific Target so

11:11

you basically take a protein crystal structure and tell your AI to design

11:16

molecules that would be very selective and active against that specific protein

11:22

Target and also could be drug-like or so something that is not a poison and can

11:28

get into the uh into the body uh the oral Administration so not injectable

11:36

for example and we've emphasized number of those molecules that's the sum uh and

11:42

published our first paper 2018 submitted 2017 uh and then in 2019 we demonstrated

11:50

a race where another company challenged us to design small molecules against a

11:57

very well known Target and fibrosis uh and we very rapidly generated small

12:03

molecules tested them against the Target and demonstrated that the generation

12:08

conditions were confirmed using experiments all right so that's 2019 and then we

12:14

productized it so we developed software around it with many many many different generative models uh for chemistry and

12:21

for biology and released it to the pharmaceutical companies and now 10 out of the top 20 pharmaceutical companies

12:27

have used their Tools in one way or another so it's actually a pretty widespread uh in drug Discovery in

12:35

longevity uh generative techniques we use them for the first time 2017 where

12:42

we used age as a generation condition so that's where you can

12:48

generate very high quality synthetic biological data for a person that you

12:55

that does not exist right so for example you can take yourself as a template say okay well I'm Christian uh here is my

13:03

gene expression profile and I want to see how this gen expression profile looks like in 20 years so you say okay

13:12

well generate according to the gene expression uh 1000 versions of Christian

13:19

uh skin for example um and show me different ages right so

13:25

from let's say you know 40 to 50. and

13:31

the system will generate a bunch of those gene expression profiles um you as a human would not be able to

13:37

interpret it but you would be able to uh trick pretty much every AI system that

13:42

is trained to predict age using gene expression right um and this approach allows you to also

13:49

identify valuable targets valuable genes uh that makes your gene expression

13:57

profile younger or older right so just by basically um moving this dial uh which is age back

14:06

and forth on several biological data types you can actually get valuable insights into what's making you older or

14:13

younger and also try to find causal relationships so

14:18

um I think the generative AI will also help quite a bit study aging we found

14:24

some Targets this way already uh and um we currently have the first drug that

14:31

we've designed uh using generative AI or a novel Target already in human clinical

14:37

trials uh so we just got the top line data from the phase one clinical studies

14:42

where we test safety now we're going into phase two for a very broad

14:47

indication with no cure um but the way we found this target was also using aging research so I think

14:54

generative AI in uh biomedicine is very well established

15:00

and very broadly used it's great I had a couple of weeks ago I

15:05

had a conversation with checks Canal he coined the term irum's law uh in 2010

15:11

and we were talking about the big problems that we have in drug Discovery and Drug development and I would like to

15:17

hear your opinion on what generative Ai and artificial intelligence can do I mean the basic principle you also said

15:23

in some interviews that 90 of drug candidates fail in clinics and when we look from uh coming from the lab bench

15:31

we are probably about 99 failure rates and amazingly the expenses for one track

15:37

in the last 10 years went up from roughly one billion dollars per track uh to three to four billion dollars

15:44

how do you see that artificial intelligence or how can artificial intelligence help to solve these

15:50

problems to make drug development faster more accurate and also cut back the

15:57

costs where is what is the role of generative AI in this field so sure thank you for asking this

16:03

question uh because I would be able to answer first in numbers right with real

16:08

facts and then give you a little bit more of a futuristic for you so as you

16:13

know our company uh we raised uh just over 400 million but mostly uh it was in

16:21

the last two years uh so from uh mostly biotechnology investors who understand

16:26

biotech uh and um last year just in 2022

16:32

we nominated nine pre-clinical candidates so eight internal and one for

16:39

for an external collaborator uh and in small molecules that's a very large

16:44

number usually a big pharmaceutical company would nominate uh you know four

16:49

five six uh sometimes when they're lucky they can go I don't know a big Pharma

16:56

company uh that nominated them for example in a single year

17:02

um it would be you usually they have a lot of programs but many of them are

17:07

external and many of them are in biologics so small molecules uh you

17:13

usually don't have the space of r d uh usually to get to this point to a

17:19

point of pre-clinical candidate you would or with all the failures you would

17:24

probably spend uh uh well is a biotech probably around 100 million dollars

17:29

right if you're going after new targets uh and uh in Pharma it's going to be

17:35

even more expensive because they they usually waste a lot of time and effort on not

17:41

uh all kinds of bureaucratic things right and just people do not focus on work as hard and fast as in

17:49

biotechnology companies so this Improvement in ex and acceleration in uh

17:56

drug Discovery is real so it's not just blah blah futuristic right so

18:01

non-proclinical candidates in a single year it's we've demonstrated that in

18:06

2021 we nominated only two pre-clinical candidates um and this year we expect to you know

18:14

hopefully exceed our internal record uh and some of these drugs went into Clinic

18:21

so usually by this time the failure rate is very high I am again we raised only

18:29

400 and we managed to um uh get to pre-clinical candidate

18:34

stage with now since we started doing our own Discovery

18:39

um we managed to progress 11. uh it's a pretty big number and you can compare

18:48

drug Discovery to space exploration right so first you need to launch your

18:53

first uh um uh first rocket right reach the orbit uh and or you know sub orbital

19:00

flights uh but still with some payload delivery uh and the most difficult one

19:06

is your first right or your second but then when you demonstrate that it

19:12

consistently works uh you can do much more and also people start believing in

19:18

it so I think generative AI already transformed drug Discovery and we have

19:24

contributed significantly to this process because we also in addition to

19:29

uh using generative AI internally we also give it to other people in the form of software

19:35

so now a lot of people are using our software uh for Target Discovery it's so easy to use that high school students

19:41

can use it and they do uh for small molecule generation uh it

19:47

is not as easy to use and we only deploy it within the karma companies and but they those that understand the value

19:54

instead of you know try and say oh we are we we can do a better job uh by the

19:59

way if some some of them say that they can also Implement some of their models

20:04

in our software so it's so flexible uh but many of them are afraid of doing that because then um if you don't under

20:12

outperform well you would have to rely on our software uh

20:18

usually it's a good good idea to do both uh and um

20:23

I think that the cost component and the time component is not as important as

20:31

the probability of success component right because when you are using generative AI for

20:36

both Target Discovery and small molecule chemistry I think that you can uh

20:42

significantly improve the probability of success because very often you would go after maybe not so hard targets but when

20:50

you have a choice of a Target uh and you can very rapidly fast forward

20:56

um you can find those combos uh that have not been previously explored uh

21:03

because either people thought that it's maybe you know uh not a very competitive Target or it's a very difficult Target

21:11

or this target might not address a substantial patient population or it's

21:17

just too new so they try to ignore this Target because uh there is very little

21:22

biology and biological evidence available around this Target and we managed to match make uh the target with

21:30

the generated molecule in the most efficient way we also try to forecast

21:35

them in the future and see which targets are going to be hot because many of them are made for sale for big Pharma

21:43

companies right so we would not be able to progress many of them on ourselves

21:48

into phase three for example right it just we won't be able to is so much Capital

21:53

uh and when you are trying to do those three steps with generative AI I think

22:00

that you are significantly improving the probability of success even from the

22:05

strategy perspective because many of those uh programs within big Pharma companies they fail because of strategic

22:11

choice so they actually just decided to Pivot uh their strategic Direction and refocus

22:18

for example from oncology to CNS uh or from you know one target Discovery

22:24

philosophy to another they want to say okay we're gonna go only for uh

22:29

genomically valid targets and then you know 70 of the pipeline is stopped or

22:34

you know cut uh and they just refocus and a lot of that is uh wasted due to

22:42

um strategic reasons so when you also try to extrapolate uh the kind of hotness of the target space into the

22:49

future and try to see what is going to be valuable for big Pharma companies from the Strategic perspective I you can

22:58

also minimize that risk so I would be uh guessing right now but I think that we

23:04

can at least improve the probability of success by 50 and that's more more

23:10

important than time and cost and that's at least right because for us internally pretty much every program we

23:17

tried uh has succeeded so we failed a few times but more than 90 of the time we

23:24

succeeded to get to pre-clinical candidate and uh into the clinic at 50 the probability of success from in which

23:31

stage are we talking about clinical uh trials or are we already earlier so I'm

23:37

talking about earlier much earlier so from Target Discovery stage all the way to let's say phase one clinical trials

23:44

um and the reason I can say that is because currently we have just a few of

23:49

those programs right and uh uh so far the probability of success was almost 100 right so

23:57

um 50 reduction in probability of success uh in probably a failure is

24:03

um my personal conservative estimate and of course we did it on a tiny fraction

24:08

of the cost uh and in a fraction of time so in two years uh well two and a half

24:14

years 11 preclinical candidates it's a pretty cool number that's um that's really amazing so for

24:22

my understanding um the numbers that I have in mind are that uh when we look earlier before

24:27

pre-clinics we have a success rate of one percent and uh your programs have a

24:32

success rate of hundred percent basically an endosoty right so it means that

24:37

you beat everything out that doesn't work very quickly and move only those

24:43

compounds forwards into clinical development and clinical development that have a real high probability of

24:48

success already this is the right understanding yeah well that's the that that's I would say a little bit more

24:54

ambitious way to explain it uh but um why don't I visualize uh and share a

25:02

screen I just need to have uh screen sharing privileges um uh and I'll just show your

25:11

um followers uh one slide which is very important

25:18

so here you go um that is the slide which describes this from Stephen Paul's paper

25:26

um uh formerly at La Lily when he published it uh in nature of use drug Discovery very famous Journal uh which

25:33

describes uh uh these different steps of drug Discovery and development uh and on

25:38

his paper he described just those steps from Target to hit all the way to submission and launch

25:45

um and show that it costs about 1.8 billion dollars right from Target to hit

25:50

and here are probabilities of success on the top here is the number of years it

25:55

took uh it usually takes and here is on average uh the number of millions of

26:01

dollars that every step costs uh I added my own estimates to the slide disease

26:08

hypothesis and Target Discovery that's usually performed in Academia not in big pharmaceutical companies they very

26:14

rarely discovered with targets that get progressed into um late stage clinical trials and here

26:20

are the probability of success I estimated to be one to five percent predominantly because

26:25

um currently the NIH National Institute of Health uh their budget annual is 45

26:31

billion dollars right so they actually spent quite a bit of money but there are very few good targets that reach the

26:39

clinic and usually it takes one to ten years uh could cost you billions of

26:44

dollars for example Alzheimer's we still don't know a single good Target that

26:50

would work for everybody right uh and the disease hypothesis is poorly understood that's why we have to study

26:57

aging to understand Alzheimer's and Alzheimer's understanding may help us

27:03

understand aging a little bit more um so what we have done so far is we've

27:09

taken one of our programs from Discovery to phase one Topline data received right

27:14

so basically we can refer to it uh as phase one complete

27:20

um and we've done all this for for this program we haven't failed yet so fingers

27:29

crossed uh we did it on under three years right so usually it would take you

27:35

on you know five and a half six years sometimes a decade

27:40

uh and it would take you a lot of money right so basically close to a billion uh

27:48

we have done that on a fraction of the cost so to reach the pre-clinical candidate stage at that time

27:55

um just for idiopathic pulmonary fibrosis application uh indication it

28:01

took us about three three million dollars um of course we have tested it for many

28:06

other indications right so we're taking it also for chronic kidney disease that

28:12

specific Target uh and uh um we're doing indication expansions into skin fibrosis

28:18

as well as one fibrotic target um but just that is a pretty cool number

28:24

uh and let me see if I can um show some other interesting slides

28:30

just to explain uh this process

28:35

um and again I will have to apologize one more time to your uh listeners for

28:40

the background noise because I'm at the world government Summit um so that is how long it took us to get to

28:49

pre-clinical candidate stage uh with our anti-fibrotic uh and again this is the traditional approach this is our

28:55

approach so under 18 months under three million dollars for ipf only uh and then

29:03

um under 30 months uh into close to phase one complete when I made

29:09

this slides now we are we're here um and again we did it at the fraction

29:15

of the cost so why knowledge is the first AI generated drug to reach uh human

29:21

clinical trials and uh uh go through the phase one clinical trials

29:27

um so we made a little bit of a history hopefully congratulations uh and and and

29:33

and and uh even if we you know do not succeed in phase two it was a very

29:39

um uh cost effective Journey compared to everything else uh but we of course are

29:46

aiming for uh major success uh and uh so

29:51

this is the first drug where we managed to get uh all the way to humans we actually conducted

29:57

um a phase zero study so a small study in a small number of humans eight humans

30:03

in Australia uh healthy volunteers we call it phase zero because again the target is new right so before and we're

30:09

a small company we wanted to ensure that we understand the Target and the compound distribution a little bit

30:15

better and then we started uh phase one trial uh in 80 Health volunteers just

30:21

about the top line data um and these are all slides but we are

30:27

we have progressed with our Target tax we haven't disclosed it and we have several

30:33

other very promising targets including targets for covet so for seafood like protease uh that one is ready to go

30:40

human right now as well uh and it's a very selective broad spectrum molecule so it works pretty much in every

30:48

um habit stream that we've tested against um and we also have a lot of cancer

30:54

targets uh that are specifically tailored for um licensing to make pharmaceutical

31:00

companies so we try to have very high quality data packages that will be very attractive to

31:06

um uh to those kind of companies and just to show you a case study

31:12

um that we published openly for everybody so you can go back and Trace our work

31:19

um and we try to publish uh as much of our work as possible uh open with Open

31:25

Access um so we went after ALS which is a very popular disease actually for all AI

31:31

power drug Discovery companies uh but we try to do it in the open

31:37

um it's a rare disease but there are still a lot of patients out there and very often when I speak with uh my

31:45

friends somebody knows somebody who had ALS or has ALS right now

31:50

um we decided to you to use all of our software tools pandonics chemistry for

31:56

interior and clinical that are available uh commercially right now of a Target discoveries more molecule chemistry and

32:02

prediction of clinical trials outcomes um um utilized on the omics where we

32:08

utilized massive amount of uh data available to us from public repositories

32:14

for ALS and other diseases uh we use multiple uh AI algorithms and AI models

32:21

uh to come up with Target hypotheses also got some textual evidence that some

32:27

of those targets may be implicated on ALS are or explored various drug filters of for

32:34

for exploring targeted whether the target is good for small molecules or

32:41

biologics uh whether it's commercially viable and then we partnered with the wonderful key opinion leaders in this

32:48

field married chutkovitz uh head of Neurology at uh

32:53

uh at Harvard MGH Jeffrey rostein uh head of the brains project and uh very

33:02

prominent neuroscientist neurobiologist from Hopkins uh Bilo from Cinque

33:08

University and kejron from Mayo Clinic to work with massive data sets from

33:14

anser ALS Consortium where a lot of human ALS cells where

33:22

reprogramming to Baby cells called ipscs uh and uh genetic profiles were

33:30

evaluated uh and also they were later programmed into full-scale neurons also

33:36

with uh all kinds of molecular data coming from the samples um and also we used a lot of samples

33:43

from our own databases that are properly uh integrating and annotating massive

33:49

amounts of public available available data uh run it through pandemics and found new targets and also older targets

33:57

with drugs to repurpose and then we tested those drugs very rapidly in a fly

34:02

model from Mayo Clinic uh so you know when it comes to rare diseases um and especially the diseases of the

34:09

brain there are very few uh models where you can test your hypotheses right so

34:15

you can test it in mice you can test it in human neurons you can try to grow

34:20

them or you can use a model organism like flies for example and do very quick

34:28

tests so we're very rare so you find we can find more information about this project at als.ai uh we use this fly

34:35

model from mayo and demonstrated that in uh

34:41

okay okay all killer neurons uh we're on

34:46

the G4 C230 phenotype is observed very similar to many ALS

34:54

phenotypes that are observed in humans um many of the drugs that we predicted

35:00

are many of the targets worked and rescued the phenotype

35:05

so AI can pick the targets and very rapidly demonstrate strong rescue moderate rescue and Mild rescue and a

35:12

few targets of course didn't work right um and we've published it openly uh and

35:19

some of those uh drugs some of those targets they do have already existing

35:25

drugs uh and we've published those for repurposing I know that there are some

35:32

companies that are not trying those drugs and targets likely

35:38

so uh that's one of the ways to test your hypothesis very quickly and yeah

35:44

then we actually did our own chemistry and I'm gonna pause here because uh that's going to be a long story and uh

35:50

with a lot of uh chemistry terms and maybe it's going to be boring for your readers for your listeners

35:57

um but we managed to demonstrate that we can very rapidly with AI identify new

36:03

targets uh uh and validate those targets either using uh Knockouts or real

36:10

chemistry uh and for a very important disease with no cure and

36:17

unpublished rapidly right so we didn't even um think about commercializing this uh

36:23

when we publish uh and from any other diseases we do something similar uh but we do it internally in the in the

36:32

first slides if I remember try to set that you cut back the time that you need to bring a drug candidate into the

36:39

clinics and uh instead of any remember right it was about 414 million in a

36:44

traditional way it's uh a company needs to move a company in Phase a track candidate interface one and with your

36:51

technology you did it for three million dollars is this uh did I remember that right that's correct wow so we managed

36:58

to get to pre-clinical candidate States just for ipf indication because uh there

37:03

you need to to to claim a preclinical candidate in our world uh you need to

37:08

demonstrate efficacy in several models so at least three uh and uh um we used

37:16

one vitro model and two mice models so to demonstrate efficacy in

37:23

uh for for idiopathic pulmonary fibrosis for other types of fibrosis we of course

37:29

did many many other experiments that increase the cost pretty dramatically yeah so but for uh proving what

37:37

artificial intelligence can do interact together when they think it's a good example because you cut back the cost to one percent or you can uh 200 tracks at

37:46

the same price of one with the traditional way so it's basically it's a very amazing fee uh fit what was the

37:53

response from regulator regulatory authorities when you approach them with your artificial intelligence approach

38:00

was it a no-brainer for them and say okay it's called move forward or did you have some challenges to solve

38:08

so of course we got some interest from The Regulators especially in Asia so uh

38:15

in China the nmpa was very interested in our uh results and uh we're in active

38:22

discussions uh and uh in the US uh

38:28

we we got some interest so I actually did gave a talk uh um uh virtual talk

38:34

could be uh to the FDA at one of the big departments uh and we are trying to

38:40

establish this connection I am basically they are our program one

38:45

which is uh um going human clinical trials phase two is going to be the

38:50

highlight right so that's what we need to focus on right now

38:56

um and uh have to tool that we really want The Regulators to use it's called

39:02

in clinical which predicts the outcomes of phase two to phase three Transitions and also interpret some of the um uh

39:10

likely failures and successes right uh so we're trying to

39:15

um make this tool available to The Regulators as well currently we're selling this to

39:21

um uh hedge funds and Banks to the financial industry and also piloting with them to improve the tool

39:29

um because of course uh the industry that gets uh hurt the most

39:36

are by failure is is the other VCS and

39:42

and hedge funds that are investing in those uh uh and banks that are investing in in the companies uh that are you know

39:50

running phase two clinical trials that's where most of the failures are uh and

39:55

they usually either win a lot or or lose a lot right so um uh the ability to very rapidly

40:04

um predict success or failure and also interpret I think that's uh that's key

40:09

to the financial industry but also to The Regulators because very often clinical Pro or phase two's actually

40:16

succeed but later they fail in phase three because the clinical study was mismanaged

40:23

somewhere and our tools uh again hopefully will be able to

40:28

allow to identify those mismanaged mismanagements yeah what do you think in future is it

40:35

possible to completely um avoid clinical trials to just predict

40:41

the success with artificial intelligence and not to human trials anymore uh unless we're in phase three or uh is it

40:48

uh just uh now you won't be able to do that you need to do click human clinical trials

40:53

for sure yeah uh even uh with something so right now

40:59

there are pretty much no medicines that work 100 right and uh

41:09

many of the medicines that are that you think work they actually work for the

41:16

wrong reason so I think that

41:21

until we figure out how to simulate this entire world at the atomic level and

41:29

also recreate it with many many different people we will require human clinical trials to

41:36

be conducted I just hope that the regulatory authorities will become much

41:41

more aware of uh generative of the potential of generative Ai and actually

41:46

recent achievements and start paying closer attention to

41:52

um what we do for example uh and try to get more involved and try to accelerate

41:57

right so I think that currently the regulator is to play a much more active

42:03

role at accelerating the pace of technological progress and very often we should actually think

42:11

about how to make really rapid pilot runs where you try to set records

42:19

in certain areas right so if I were a position right now or focused on on

42:27

you know improving the pharmaceutical industry I would try to identify few

42:35

opportunities for example in rare diseases where AI could be used end to

42:41

end and I would try to partner with a few companies well for example like our

42:46

hours or with hours to try to go from Discovery to uh the clinic in the

42:55

shortest amount of time possible so you basically design a strategy you

43:00

said okay well this is the disease I want to go after preferably with with

43:05

the Target that is not obvious uh for example like ALS o and then give us green light to

43:14

conduct rapid clinical studies with novel small molecules and observe how

43:22

are those studies pan out and also allow us to get the approval as quickly as

43:30

possible right and for example if you can tangibly demonstrate that you can do

43:36

this within you know four years for example well you just cut down the time

43:43

by you know 60 percent and demonstrate you can demonstrate that

43:48

the effective drug can go into uh humans and I'm talking about to the market in

43:56

in a certain amount of time and then we can Benchmark against that pilot so you

44:01

need to have a few of those uh you know going to the Moon uh cases preferably in

44:09

a smaller scale for more rare diseases but with poorly understood biology

44:14

on in the race format just to get uh better

44:22

at uh um just to get better at uh you know

44:28

regulating for bigger bigger diseases right so I would try to do that

44:34

and currently I don't know of any politician or regulatory Authority that

44:41

would be looking at this at scale now it's interesting I mean uh with

44:46

large-scale studies in fields like covet for example are 40 000 patients uh forty

44:51

thousand people versus expand I was involved in two companies who did

44:57

research in real diseases and the timelines in clinics were just uh not

45:03

very attractive to buses I mean when I approach this season say look we need five to ten years to recruit enough patients

45:10

um they are not very happy with that timeline and uh are reluctance to invest where do you see artificial intelligence

45:16

in in the field of clinical trial to really speed up the I mean when you don't fancy patients what can you do

45:23

with that in artificial intelligence so sure well again if you go for China

45:28

for example some of the disease rare diseases are not so rare so you can do

45:35

faster Recruitment and I actually think that uh for this kind of exercise

45:40

uh in AI uh that's one of the areas where International collaboration could

45:47

help a lot right so currently we see a lot of uh what we call you know trade

45:52

Wars uh between U.S and China specifically uh and actual real wars uh

46:01

where Russia invaded Ukraine for example um but with the uh

46:09

with rare diseases and races for AI part you know AI part drug Discovery I hope

46:15

that authorities could collaborate and maybe we could recruit some of those patients much quicker

46:21

um and also I'm pretty sure if the um you know an agency like the FDA would

46:28

uh prioritize uh our feel uh in a bigger way and specifically uh you know look at

46:37

how to um uh at how to conduct such a such an experiment right

46:45

I'm pretty sure we could come to a conclusion very quickly would require just one day of a seminar uh or or of

46:52

meetings workshops to come up with a product proper strategy

46:57

um I'm pretty sure that we can also get some significant financial backing from the private sector because uh if um uh

47:06

the the regulator is would demonstrate that at the end there is some potential

47:12

uh profit involved right so uh the VCS

47:18

who would be funding something like this would have a big payout plus a seat at

47:23

the table for future uh races like that I'm pretty sure we would be able to do

47:29

that but currently nobody is motivated enough to uh to go after those

47:37

races of course the covet case was probably the best demonstration of uh

47:43

abilities and also drawbacks and inefficiencies in the industry so even

47:49

the first drug that has been properly approved uh

47:54

so why is a drug um it's still a very old molecule that

48:00

is very rapidly repackaged uh as an oral

48:06

um as an oral c3-like protease inhibitor with another drug together that helps

48:12

with you know sip inhibition with better um stability of the molecule uh so it's

48:18

definitely an imperfect uh drug that probably wouldn't go through in a normal

48:25

environment uh but during covet you've seen that they've managed to basically conduct all

48:31

the clinical work in a year and a half right so that is something that

48:39

The Regulators are capable of so now let's try and do that with a noble Target Noble molecule right so why don't

48:46

we uh try to increase the novelty and do another race

48:51

not wait for another pandemic yeah I know the pandemic was amazing I mean to see the whole industry moves so

48:58

quickly we have one question from the audience that will uh allow M kayash I

49:05

have speak the name right maybe to ask the question himself because you have

49:10

the micro um would you like to ask your question hi

49:16

I'm really enjoying talking I am azamphease from Dubai I think thank you very much I'm really I

49:24

am proud of you what you are doing uh this is I asked you before but I want

49:30

to know after I listen to your lecture uh artificial intelligence microbiome based

49:38

therapeutic tools can we apply what you are talking on the

49:44

microbiome new molecule so absolutely so if you have a very good

49:52

microbial Target uh so expressed by bacteria uh Rai can

50:01

perform as well on this target as it does on human targets as long as you

50:07

have the crystal or uh you have a predictive crystal structure from alpha

50:14

fold or reset fold or some other algorithm um we could do similar things on

50:20

microbial targets plus um uh we could look at new Target

50:26

Discovery uh utilizing very similar tools like pandemics and uh what we did

50:34

specifically um so about 2017 I trained the Deep

50:40

neural network uh to predict human chronological age in a healthy State using uh microbiomic data using whole

50:49

genome sequencing data um of human gut microbiome uh and we got

50:54

to reasonable accuracy for prediction of uh human age again that's one of the features that is present on pretty much

51:01

every data set right uh and we demonstrated which bacteria are more

51:07

senopositive and xenonegative so contribute to you quote unquote looking

51:13

younger to the Deep neural network or quote-unquote looking older to the Deep neural network uh and that showed you

51:21

that deep neural networks can actually zoom into specific bacterial species uh

51:28

that may perform certain uh certain roles play certain roles in different

51:36

biological processes so in this case in aging um they are of course the causalities

51:42

poorly understood but we can at least narrow down to specific by uh specific

51:47

microbial species um and a young chemistry is chemistry so

51:53

uh in chemistry we can develop small molecules uh discover small molecules

51:58

for bacterial targets no problem uh for

52:04

for disease Target research we can possibly a narrow down to individual

52:10

bacterial species and then within bacterial species uh we could identify possible targets just by looking at what

52:17

uh um proteins are important for the various uh biological processes within

52:23

bacterial species so there are many ways to integrate AI in drug Discovery using microbiome uh

52:32

and then there are many companies uh in addition to ours that are actually doing

52:37

maybe even a better job at Target Discovery using microbiome I personally

52:42

do not prioritize microbiome as a data

52:48

type within on silica because we have many many other data types to focus on

52:54

um is of course promising but I think again for from the drug

53:01

Discovery perspective currently we are prioritizing other data types so we like

53:06

to have for human proteins as targets thank you thank you very much I am

53:12

looking to work with you or under your umbrella thank you very much well done thank you would love to explore any

53:18

collaborations thank you talking about collaborations uh you

53:24

mentioned that your company operates in the Middle East in China and I also

53:31

think in the United States and in Canada uh why did you decide to go this way

53:38

so we try to go uh after the best talent in the world very often we hire through

53:45

competitions uh and in Europe for example we used to hire through massive

53:51

native National competitions hackathons um and then uh for example when Russia

53:58

invaded Ukraine we realized that much of this valuable Talent from Europe could

54:04

be moved to the Middle East so I managed to get quite a bit of the quote-unquote

54:10

AI refugees um originally into Abu Dhabi and then we

54:15

realized that that's a great place to actually uh host Talent specifically I

54:22

talent and we started hiring and we grew the team uh to about 40 uh here

54:32

um and that's one of the reasons for international expansions expansion into the Middle East uh China is very special

54:39

case so much of the biomedical research and contract research infrastructure is

54:46

in China so they are over the past 20 years they invested over half a trillion dollars into biomedical

54:53

infrastructure and research infrastructure so you don't need to own your own Labs you can actually hire our

55:01

contract research organizations to do it for you as and and for an AI focused

55:07

company uh not the the absence of this uh

55:14

Labs uh is actually a competitive Advantage so you have to if you want to

55:21

move quickly you can you can actually move very quickly by Outsourcing many

55:27

many different experiments to many different labs do it in parallel and

55:32

also have redundancy right so very often one lab may screw it up another lab May

55:37

succeed which one do you go with right so then you repeat the experiments in both very often if you do it internally

55:45

and if you succeed um in places where you should not succeed uh and you proceed with the ex

55:52

with the program uh it will be very damaging in later stages so we try to

55:57

introduce quite a bit of redundancy when we are talking about uh um experimental data so China is very

56:05

important it's very important place to do experiments uh and uh if you do have

56:10

resources there overseeing um the large number of full-time

56:16

equivalents within contract research organizations you can get higher quality

56:22

and faster speed and even uh create an ecosystem of those contract research

56:28

organizations that is very seamless and is better than your than having your own lab of course we do have our own fully

56:34

robotics laboratory as well but uh relying on those cross is key and of

56:42

course Montreal Canada is a great place to hire AI Talent that's the capital of

56:48

artificial intelligence one of the capitals the second one would be Toronto

56:53

um so still Canada so proud to be Canadian and

56:59

um uh and uh Taiwan for example great part

57:05

of China where um you can hire a really amazing computational chemists and in Hong Kong

57:12

we do have our Target Discovery Group which um utilizes a lot of REI tools to

57:17

discover new targets but actually it's really unusual to have this geographic distribution but for us it works and it

57:24

makes us a much more resilient organization and I think that into this level of diversity helps us innovate at

57:32

a broader scale and you can make use of the best from all countries so it's

57:39

pretty amazing it's pretty amazing the approach that you have it's I like it you mentioned before you have fully

57:45

intelligent robotics lab can you talk a little bit more about that

57:51

so sure uh I already tested if I do show you the video uh via Zoom it's not going

57:57

to be of good quality and it's gonna lag so I'll just talk through it and maybe we can post a link to this link in

58:04

comments uh so we we realized that now that we

58:09

have demonstrated that AI can very efficiently identify new targets that

58:15

actually get validated using real experiments we can now do it at a higher

58:20

scale at a broader scale and we wanted to automate this process of Target

58:26

Discovery so we are not reliant on this incremental data for our Target Discovery but we can

58:33

um now perform experimental validation of our algorithms much faster so we have

58:39

a lab where you can throw in a sample imagine that a sample comes from some

58:44

animal model of disease of disease right so you take a piece of animal tissue I

58:51

just don't want to use the human analogy um where you've got for example certain

58:57

cancer of a specific tissue you throw it into the robot the robot picks it up uh

59:03

grinds it micro plates it does quality control analysis then sends it

59:08

automatically into another room where part of the sample is being placed on the incubator

59:14

uh so you grow some of those cells uh or even primary tissue

59:20

um or you store it uh and but part of the tissue goes into the

59:25

um high resolution imaging system so let's get image I also

59:31

um incubate uh some of this tissue with contrast markers so you get high content

59:37

uh Imaging with uh fluorescence so part of the sample gets destroyed uh when it

59:44

gets sent to another room or where you get uh Onyx data so transcriptomics and

59:51

methylation data and several other data types all this data goes back into ai ai

59:57

decides which targets are important in that specific sample and which uh drugs already work

1:00:07

on this target fix those drugs from the libraries uh and test them in a hype of

1:00:14

Facebook with manner are different incubation times different concentrations and then every sample

1:00:21

goes through the same exact uh cycle so it gets imaged or it gets well high

1:00:27

resolution and also fluorescence Imaging and you get epigenetic data and you get

1:00:36

gene expression data and plus other data types and you then you compare your

1:00:42

predictions to the actual experimental data and see if the target if the targets

1:00:48

were found and if the compounds were and in parallel you also try to

1:00:54

find some normal biology so find some normal biological hypotheses that might

1:00:59

be valuable in the future so this lab is more of a Target

1:01:05

Discovery plus Target validation lab now we are uh miniaturizing it we're trying

1:01:11

to make it smaller so you can even do this at the hospital level imagine if you could do personalized drug Discovery

1:01:18

for individual patients well uh or perform some uh already

1:01:23

um established personalized medicine services or um decision support services for

1:01:29

Physicians and at the same time discover targets so you could place those labs in

1:01:35

regions that never even aspire to be a player in biotechnology um how small how smart can you make it I

1:01:43

think I can do it in two rooms in two rooms this is practically you can can do

1:01:48

it everywhere then it's not yeah currently is the entire floor which is automated

1:01:55

um and it's a pretty complex system so it's many rooms uh performing different tasks uh with many different types of

1:02:03

equipment I think if I try to place this equipment in 3D and also

1:02:10

make some of my own equipment uh I would be able to automate this

1:02:16

process with just tools so it's possible to run the processes in

1:02:22

every Hospital of the world basically with uh well every major one that can afford something like that yeah or if

1:02:29

there is a government program that allows you to uh uh to do something like that so

1:02:36

uh currently that's a dream right but and I I don't I cannot guarantee you

1:02:43

that we will have a hospital-based robot but you know three years ago I did not

1:02:50

even dream about the lab that we have yeah this would be my final part for

1:02:57

this conversation so looking into the future uh but before I ask that we're coming to the end of the podcast uh is

1:03:03

there any question open that you would like me to ask is there anything that I missed that you think is important to

1:03:09

share well I think that you covered pretty much everything I must say that the most

1:03:16

important disease that we have are in the world is actually aging because

1:03:22

we will all lose function and die and in this process we will also get a

1:03:30

bunch of diseases and Trauma and uh it's never a happy ending

1:03:37

so it's always lots of function so I think that we need to look at

1:03:44

um and prioritize Therapeutics that have dual purpose that can treat a disease but at the same time

1:03:50

Target um important uh biological processes that are implicated on Aging

1:03:56

and what we do at the Silicon what I do are in general

1:04:02

um is looking for those Dual Purpose Therapeutics so I think being Dual

1:04:08

Purpose aging and disease that's truly key for a identifying Blockbuster drugs

1:04:14

because uh the probability that your drug is going to work for many people at

1:04:20

the same time is going to be higher if you are addressing a very broad biological process

1:04:27

um and the second we need to figure out aging right because that's that's the main killer so

1:04:34

this year 60 million people are gonna die of due to aging that's more than any

1:04:40

war or any cataclysmic event or any plague

1:04:46

um and we need to figure out how to increase health span and significantly

1:04:53

delay definitely do you do you think it's possible uh to

1:05:00

delay definitely there's this uh there was a science fiction book that I read I think 40 years ago

1:05:05

um where the Assumption was that one day uh some Pharma industry in an alternative reality will develop a drug

1:05:13

that basically makes death a conscious decision so people don't age people don't get sick anymore do you believe

1:05:19

it's possible to come to that point in our reality somewhere in the future

1:05:25

so with pharmacological means I'm talking about small molecule drugs or biologics uh or uh even RNA Therapeutics

1:05:34

you won't be able to dramatically increase lifespan so I'm talking about

1:05:39

maybe you know plus minus plus uh 40 50 that's probably possible if you combine

1:05:46

a bunch of Therapeutics that a intervene with your metabolism and be

1:05:53

um uh improve certain repair functions uh and maybe even

1:06:00

um stimulate certain uh certain biological processes that are just too

1:06:05

expensive for the body to run all the time or where we do not have a biological program

1:06:11

um uh so I think that significant life life

1:06:17

extension is possible using pharmacological means but not dramatic to go dramatic you need to look past

1:06:26

pharmacology it's just pharmacology is the most uh credible way to uh extend lifespan and

1:06:35

um it's also something that is very sustainable so if you are very good at

1:06:42

discovering and delivering novel Therapeutics for diseases uh and you are also prioritizing aging

1:06:50

as your primary uh outcome measure so the effects on uh on on

1:06:56

the health span and lifespan of your patients um I think that this is the area where

1:07:04

you can achieve commercial success uh and then fund other areas of

1:07:11

longevity research that may give you uh more longevity dividend compared to just

1:07:17

from Pharmaceuticals yeah I think there's a lot of Lifestyle um from a particulars one way lifestyle

1:07:24

is another way and probably with Elon Musk research or research that's

1:07:29

going in that direction maybe one day it's possible to upload Consciousness into into the account and download it to

1:07:36

an artificial intelligence Auto in a toy robot yeah I wouldn't bet on that

1:07:41

uh but uh um it's definitely a very promising area of research for other purposes also

1:07:49

aging and Alzheimer's CMS diseases uh but yeah I could look at uh maybe

1:07:56

gradual replacement of neuronal tissue uh as the uh you know way to rejuvenate

1:08:03

the brain but uploading that's something that uh we may not be able to master I mean

1:08:10

creating a copy yes but is it going to be you most likely not why not create a

1:08:15

clone then right yes yeah that's a big question I have one final question left when we are talking about the future how

1:08:22

do you envision generative and Robotics impacting the future of healthcare and Medicine

1:08:29

well I think it's already impacted in a massive way uh so and partly uh due to

1:08:35

our work in generative chemistry and generative biology in uh uh you know

1:08:40

2016-2017 2018 so seminal work uh and now it has propagated uh so I think the

1:08:48

generative AI is going to transform our life uh beyond recognition in every way right so we are actually moving into the

1:08:56

new uh era of Knowledge Management so instead of retrieving knowledge uh

1:09:02

through search and through by browsing the different directories we can now uh

1:09:08

generate uh knowledge and complete response to your queries with reasonable

1:09:16

accuracy this accuracy is expected to improve over time uh and you would be

1:09:21

able to get purpose-built systems for Discovery so

1:09:27

just like our Ponder omics platform but even more reliant on generative uh AI

1:09:33

with much better user experience and

1:09:38

perhaps conversational AI um and yeah I think that we're getting into

1:09:45

the area where generative AI will play a role in pretty much every area of drug Discovery and development most important

1:09:51

uh area where I see the application of generative AI is the ability to generate

1:09:57

high quality synthetic patient data right so currently people are just too

1:10:03

sensitive about their Healthcare data being misused for all kinds of purposes

1:10:08

by the way I do not know many cases where people were hurt by having their Health Data accessed they are hurt by

1:10:16

having their financial records for example accessed or you know credit confirmation identity information about

1:10:23

health care information very rarely there are a lot of hypotheses but insurance companies are

1:10:30

um regulated right and prohibited from using genomic information for

1:10:36

discrimination so for drug Discovery I think we should be able to use this information

1:10:42

um however since people are so afraid of sharing their biological data we can now

1:10:49

use AI to generate high quality data instead of taking them from taking it from specific individual just by like by

1:10:56

asking AI to generate specific biological data sets with desired properties are trained on publicly

1:11:04

available data that's that's a great future that's a great future Alex thank you very much

1:11:11

for your time uh and uh on the world's

1:11:16

uh government Summit and uh I wish you and your team all the best for the

1:11:22

future uh you're doing an amazing work and I believe you your team will move

1:11:27

the healthcare industry globally forward and enjoy the summit well thank you Christian I don't

1:11:33

apologize to you and your listeners uh

1:11:39

and people who watch this uh for the background noise and also for the kind

1:11:44

of presentation from the field um I just couldn't change it and I couldn't escape so uh thank you for

1:11:51

bearing with me it's a nice it gives a nice flavor to the episode I like it I like it thank you very much Alex and have a great day

1:11:58

thank you bye bye you bye-bye thanks for joining me on this journey to discover

1:12:04

the groundbreaking potential of artificial intelligence in drag Discovery and anti-aging research if

1:12:13

this episode has blown your mind and opened your eyes to the incredible power of generative AI then smash that like

1:12:21

button and hit subscribe to never miss out on future episodes together we can spread the words and

1:12:29

inspire a revolution in the pharmaceutical industry that we change the world as we know it so don't be shy

1:12:36

leave a comment and share this video with everyone you know to help us reach

1:12:43

more people and make a real difference thanks for being a part of the movement

1:12:50

towards the brighter healthier and longer future

1:12:55

enjoy your day