The medical potential of AI and metabolites | Leila Pirhaji

Prevoditelj: Ivan Nekić Recezent: Sanda L

In 2003,

2003. godine,

when we sequenced the human genome,

kad smo razložili ljudski genom,

we thought we would have the answer to treat many diseases.

mislili smo da ćemo imati odgovor za liječenje mnogih bolesti.

But the reality is far from that,

No, stvarnost je daleko od toga,

because in addition to our genes,

jer osim naših gena,

our environment and lifestyle could have a significant role

naše okruženje i način života mogu imati značajnu ulogu

in developing many major diseases.

u razvoju mnogih velikih bolesti.

One example is fatty liver disease,

Jedan primjer je bolest masne jetre,

which is affecting over 20 percent of the population globally,

koja pogađa preko 20% stanovnika svijeta,

and it has no treatment and leads to liver cancer

i nema joj lijeka, a vodi do raka jetre

or liver failure.

ili zatajenja jetre.

So sequencing DNA alone doesn't give us enough information

Dakle, sekvenciranje DNK samo po sebi ne daje nam dovoljno informacija

to find effective therapeutics.

za pronalazak učinkovitih terapija.

On the bright side, there are many other molecules in our body.

Dobro je što postoje mnoge druge molekule u našem tijelu.

In fact, there are over 100,000 metabolites.

Zaista, postoji preko 100.000 metabolita.

Metabolites are any molecule that is supersmall in their size.

Metaboliti su bilo koja molekula supermale veličine.

Known examples are glucose, fructose, fats, cholesterol --

Poznati primjeri su glukoza, fruktoza, masti, kolesterol --

things we hear all the time.

ono o čemu stalno slušamo.

Metabolites are involved in our metabolism.

Metaboliti su uključeni u naš metabolizam.

They are also downstream of DNA,

Oni su na nižoj razini od DNK

so they carry information from both our genes as well as lifestyle.

pa nose informacije iz naših gena, kao i stila života.

Understanding metabolites is essential to find treatments for many diseases.

Razumijevanje metabolita je ključno za pronalazak tretmana za mnoge bolesti.

I've always wanted to treat patients.

Oduvijek sam željela liječiti pacijente.

Despite that, 15 years ago, I left medical school,

Unatoč tome, prije 15 godina, napustila sam medicinsku školu

as I missed mathematics.

jer mi je nedostajala matematika.

Soon after, I found the coolest thing:

Ubrzo potom otkrila sam sjajnu stvar:

I can use mathematics to study medicine.

Mogu koristiti matematiku za studij medicine.

Since then, I've been developing algorithms to analyze biological data.

Od tada razvijam algoritme za analizu bioloških podataka.

So, it sounded easy:

Dakle, zvučalo je jednostavno:

let's collect data from all the metabolites in our body,

prikupimo podatke o svim metabolitima u našem tijelu,

develop mathematical models to describe how they are changed in a disease

razvijmo matematičke modele za opisivanje kako se mijenjaju u bolesti

and intervene in those changes to treat them.

i intervenirajmo u te promjene kako bismo ih liječili.

Then I realized why no one has done this before:

Tada sam shvatila zašto to nitko nije učinio prije:

it's extremely difficult.

to je iznimno teško.

(Laughter)

(Smijeh)

There are many metabolites in our body.

Postoje mnogi metaboliti u našem tijelu.

Each one is different from the other one.

Svaki od njih različit je od onog drugog.

For some metabolites, we can measure their molecular mass

Nekim metabolitima možemo mjeriti molekularnu masu

using mass spectrometry instruments.

instrumentima za spektrometriju mase.

But because there could be, like, 10 molecules with the exact same mass,

No kako bi moglo biti, recimo, 10 molekula s istom masom,

we don't know exactly what they are,

ne znamo točno koje su,

and if you want to clearly identify all of them,

pa ako ih želite sve jasno identificirati,

you have to do more experiments, which could take decades

treba raditi još eksperimenata, što bi moglo trajati desetljećima

and billions of dollars.

i stajati milijarde dolara.

So we developed an artificial intelligence, or AI, platform, to do that.

Tako smo razvili umjetnu inteligenciju, ili AI, kao platformu koja će to učiniti.

We leveraged the growth of biological data

Iskoristili smo rast bioloških podataka

and built a database of any existing information about metabolites

i izgradili bazu podataka svih postojećih informacija o metabolitima

and their interactions with other molecules.

i interakcija njih s drugim molekulama.

We combined all this data as a meganetwork.

Povezali smo sve te podatke u megamrežu.

Then, from tissues or blood of patients,

Zatim iz tkiva ili krvi bolesnika

we measure masses of metabolites

mjerimo mase metabolita

and find the masses that are changed in a disease.

i tražimo one koje se mijenjaju u bolesti.

But, as I mentioned earlier, we don't know exactly what they are.

Ali, kao što sam spomenula ranije, ne znamo točno koji su.

A molecular mass of 180 could be either the glucose, galactose or fructose.

Molekulska masa 180 može biti glukoza, galaktoza ili fruktoza.

They all have the exact same mass

Sve one imaju iste mase

but different functions in our body.

ali različite funkcije u našem tijelu.

Our AI algorithm considered all these ambiguities.

Naš AI algoritam uzima u obzir sve te nedorečenosti.

It then mined that meganetwork

Zatim pretražuje tu megamrežu

to find how those metabolic masses are connected to each other

da vidi kako su te metaboličke mase međusobno povezane

that result in disease.

kad rezultiraju bolešću.

And because of the way they are connected,

I po načinu na koji su povezani,

then we are able to infer what each metabolite mass is,

onda možemo zaključiti što je svaka metabolička masa,

like that 180 could be glucose here,

kao, ovdje bi 180 mogla biti glukoza,

and, more importantly, to discover

i, što je još važnije, otkriti

how changes in glucose and other metabolites

kako promjene u glukozi i drugim metabolitima

lead to a disease.

dovode do bolesti.

This novel understanding of disease mechanisms

To novo razumijevanje mehanizama bolesti

then enable us to discover effective therapeutics to target that.

omogućuje nam zatim otkrivanje učinkovitih ciljanih terapija.

So we formed a start-up company to bring this technology to the market

Stoga smo osnovali start-up tvrtku kako bismo tu tehnologiju stavili na tržište

and impact people's lives.

i poboljšali živote ljudi.

Now my team and I at ReviveMed are working to discover

Sada moj tim i ja u ReviveMed radimo na otkrivanju

therapeutics for major diseases that metabolites are key drivers for,

terapija za glavne bolesti kojima su ključni pokretači metaboliti,

like fatty liver disease,

poput bolesti masne jetre,

because it is caused by accumulation of fats,

jer je uzrokovana nakupljanjem masti,

which are types of metabolites in the liver.

koje su vrste metabolita u jetri.

As I mentioned earlier, it's a huge epidemic with no treatment.

Kao što sam spomenula ranije, to je ogromna epidemija bez lijeka.

And fatty liver disease is just one example.

A bolest masne jetre je samo jedan primjer.

Moving forward, we are going to tackle hundreds of other diseases

Ubuduće ćemo se boriti sa stotinama drugih bolesti

with no treatment.

za koje nema lijeka.

And by collecting more and more data about metabolites

Prikupljanjem sve više podataka o metabolitima

and understanding how changes in metabolites

i razumijevanjem kako promjene metabolita

leads to developing diseases,

dovode do razvoja bolesti,

our algorithms will get smarter and smarter

naši algoritmi će postajati sve pametniji

to discover the right therapeutics for the right patients.

u otkrivanju pravih terapija za pravog pacijenta.

And we will get closer to reach our vision

Približit ćemo se ostvarenju naše vizije:

of saving lives with every line of code.

spašavanja života sa svakim retkom koda.

Thank you.

Hvala vam.

(Applause)

(Pljesak)