Web3 on the Verge of Saving Lives: The Power of DeSci - Wojciech Sierocki

all right H hey folks how are you doing great um so we had a complicated title of a talk and I took the liberty to modify it a little bit to make it simpler because we are going to talk science so I wanted to make it really simple um so in plain English what we are going to be talking about today is um what the hell is science what the hell is decentralized science right so uh leveraging leveraging blockchain for open collaborative and efficient um science so um to kind of give you a breakdown um of the whole landscape and how blockchain is being used uh for scientific discovery funding research and um and IP management uh first I will uh tell you a little bit more about challenges in traditional science um and then I will reflect uh about projects that are building in specific verticals uh to address the challenges in the whole uh value chain of scientific Discovery from funding of innovation later through execution uh of research and uh last but not least um uh to publishing of the results of the of the of the research um okay so let's get started if you guys happen to have any questions uh in the middle of the talk just you know raise your hand and we'll be happy to answer them as we go we know that at least I tend to forget questions uh if I don't put them down or I don't ask them immediately so you know let's make it a little bit more collaborative if you want uh um cool so for those of you who uh don't happen to be scientists who don't work um with science uh science of course as um as an old-fashioned traditional industry is Frau with many challenges so among those uh four uh that I think are the most important ones are funding bottlenecks um so difficulty in getting your research funded so you're a talented scientist you have a novel breakthrough idea uh to be able to uh put together um a study of course you need funding and there are many challenges with that um uh because majority of the research is funded uh through government grants uh second area is publication bias and I'm going to give you a couple of examples how actually publication bias uh affects everybody here in the room even though you don't know it um the third is data accessibility and reproducibility um so I'll show you that vast majority of uh research especially in biomedical Sciences is not um replicated so even though some results are are out there uh other folks are not able to take the same study with the same design and replicate the same results uh last but not least they are intellectual property issues um so of course majority of commercial research is done with hope of uh inventing something new of patenting it at the end of the day and you know gaining monetarily um monetary benefit from um from distribution of that intellectual property right uh so let's zoom in into funding um bottlenecks So currently majority of science majority of uh Discovery and research is funded by uh either government grants or um private sponsors um there is um a significant bias um with Government funding at least towards safe experiments so something that is not you know quite a moonshot not you know that exciting but it wouldn't make look a full if I decide to sponsor it right if I if I grant um if I if I give away a grant towards that idea um there's also bias towards wellestablished researchers uh so if you are a researcher affiliated with a reputable University with a bunch of Publications under your belt of course it's much easier for you to get subsequent funding um which once again uh is a bias towards safer experiments um uh at um at the at the Peril of more um uh Boulder ideas um there's a lot of time and effort um connected with obtaining government uh grants uh and of course there are also uh neglected and orphan diseases so for those of you who um happen to also be longevity enthusiasts uh there's a very good point uh made by researchers working with longevity um that you know this area is underfunded uh it's not yet uh a point of interest for major pharmaceutical companies there's not much uh commercial innov ation coming through the research pipeline therefore uh it's not uh um it's not funded adequately um and by extension uh there cannot be as much Innovation as we hope um there would be um let's look at the second one uh which is not so easy to understand yet I think it's the most important one here so let's look at the publication bias and I'm going to give you a couple of examples of some common drugs um and the research Discovery process uh behind bringing those drugs to market and using them um as you know medicines that all of us uh may be taking one day in the future I'm a medical doctor by training um and most of the drugs here I actually myself prescribed and I did it because there was peer reviewed uh literature and there was a scientific consensus that those drugs are good for the patients right so for that reason I was able to infer uh with certain logic that I can give this drug to the patient and that would be the best treatment option right not always right so for example U there is a drug called lanite which is an anti- arhythmic drug uh and uh there was a study done in 1980s uh where they took a bunch of people about aund about a 100 people and they divided them into two groups right so 50 people got the pill uh after a heart attack and 50 people uh got a sugar pill after a heart attack out of the 50 people who got the real drug 10 of them died and out of the 50 people who got a fake drug a placebo only one person died so I'm going to ask you is that a good drug not really right so for that reason this drug wasn't patented wasn't brought on the market but this research because it was a negative research it was you know nothing extraordinary nobody invented a cure for cancer it was a negative study it was a bad drug nobody wanted to hear about it it wasn't published okay and for that reason subsequent drugs also from this the same class were successfully brought into the market and were used for many many years treating um myocardial infarction patients so patients with heart attack and it's estimated that because of the use of those drugs about 100,000 people in the US Alone died over the period of 15 years why because that research wasn't published okay so we could have known before that these drugs aren't really effective uh as you know a drug to manage patients after a heart attack but there is a strong bias in Publications in research towards positive science okay so experiments that actually work uh as opposed to experiments that do not you may think that was you know a oneoff event but actually this is um emblematic of uh a much broader phenomenon in science so let's take another example reaboi uh this is a drug uh which is um an anti-depressant drug so doctors prescribe it to treat depression and mood uh disorders uh and I actually read the studies U that support the evidence for using this drug uh in patients so there was one study that compared reboxetine versus placebo and it found out that reboxetine is better than a sugar pel okay and there were three studies that compared reboxetine with other anti-depressant drugs and it found out that it's just as good as any other anti- depressant drug is it a good drug that's a question for you is it a good drug better than a sugar peel and you know on power with other drugs from the same class uh they compared it with uh three other drugs from the same class okay that were the standard of care before yeah yeah so you know that was also my conclusion and um people were using this drug for for many many years because they thought that it's better than sugar better than nothing and comparable to drugs what they did not know is that there were actually many more studies done with the drug called reox so there were seven studies uh comparing reboxetine to Placebo and there were 12 studies that were comparing reboxetine to another anti depressant drug out of seven studies six of them were negative so they showed that this particular drug is no better than a sugar peel they didn't get published only the one that showed that this is an effective drug got published out of the 12 studies that compared trotin with another anti uh depressant nine were negative so they found that it's actually worse than different treatments they didn't get published three of them were positive so you know as good as are the drugs and they got published is it a good drug exactly it's not a good drug and I felt cheated because I didn't have all the information at hand right um let's take a broader case Okay because that's only one drug in the anti-depressant category but there was a study actually done by craan um Library where they took all the trials in human subjects all the clinical trials in human subjects of anti-depressants in the US done over the period of 15 years there were 38 trials that found out that those drugs are working well and there were 36 trials that found out that those drugs are not effective okay out of 38 trials with positive results 37 37 of them were published out of 36 trials with negative results three of them got published okay so this is how science is done still up to this date so even though you have you know scientific evidence you have peer reviewed uh Multi Center double blind Placebo controlled trials you cannot always you know be um totally trusting in them and you know having absolute certainty that the choice that you're making is in fact the best one um later there was also another study done which found out that statistically positive findings are twice more more likely to get published than negative findings okay so imagine I have a coin and I flip it 100 times and then then I don't publish 50 results you may think I have a coin with two heads a double-headed coin which is not the case okay and imagine if I did a study and I took 50% of my data points and I only did a study based on 50% of the data points that would be essentially a research fraud that would be bad science that would be a bad study yet for some reason we allow you know people to do 36 trials and publish only three of them and we call it good study okay because the U the responsibility uh disseminates among funding bodies researchers government agencies sponsors hospitals and so on it's really hard to point out who's responsible for this scientific publication bias that's why nobody is held accountable okay and we have this kind of bad SES so this is what we call publication bias it's a huge huge huge huge huge problem then we have data accessibility and reproducibility bias okay so if I run an experiment I have a coin okay I toss it 100 times I have you know approximately 50 heads and 50 Tails I give the same coin to Oliver he tosses it 100 times he should have you know similar results if we repeat it over and over again yet in science especially in you know science that's really important for us for example in preclinical oncology okay so cancer research folks taking molecules and drugs and trying to find out what actually works to kill cancer cells only 11 to 20% of studies in basic biology can be replicated okay so about 80 to 90% of studies were done but just once and they can't be successful fully replicated which is of course another huge huge problem last but not least we have intellectual property issues so question replicated because most of the times the data upon which the study was done is not available so it's very hard to you know test the the whole methodology how they did it exactly and so on most of the times only the results are published okay and of course there's also a strong incentive to publish positive results so it can happen that in fact you know it's a 50/50 uh but only the positive results get published okay so we don't have the full picture that's going to be the second part of my talk so if you bear with me I'll just you know briefly go through the intellectual property issues um so essentially um for those of you who know how intellectual appri works you invent something you publish it you patent it you have something of value which is called a patent then you can sell it and that's how um commercial research companies make money there are a couple of challenges though uh so for example if a group of independent thinkers inside a desent cized autonomous organization spread all over the world wanted to invent something it's actually very hard to fractionalize this uh intellectual property distributed among many people who contributed to This research okay and for that reason um people say that it Al also stifles uh research cooperation and makes um cooperation you know less uh probable to happen so uh here comes dii all dressed in white and dii is actually based on man principles of uh open science so research monetization we should all be able to you know participate in research um uh and in Discovery and be able to be rewarded uh diverse funding not only from government agencies and Pharma but also you know from people who actually care about research in specific Area Community involve involvement open sourcing so publishing of the data publishing of the studies also the negative ones upon which uh our research was done an onchain record of all these operations so you can actually go and verify what was going on how the research was done so those are kind of the main assumptions of decentralized science this Builds on top of uh another movement which is called open science which is quite old um and kind of more traditional but as you see um it um it actually propagates same principles so reproducibility of results scientific Integrity citizen science which is similar to community involvement promotion of collaborative work um ease of access to knowledge for all and stimulation of innovation so uh because this is a pretty new space uh folk started building in decentralized science only a couple years ago um I will actually be able to give you a good overview of what's going on in the space for the remaining seven minutes if I hurry up um so if you look at scientific research uh the way it's done first you need funding then you need to actually do some stuff uh you know execute your research and then at the other end of the funnel you hope to publish the results you know become a professor and live in glory for the rest of your life so this is kind of you know value chain of decentralized science or value chain of science Asis uh the first project uh which I think is worth keeping an ion which is kind of an OG in the decentralized science space is a project called molecule um they have um a platform technology for tokenization of Ip uh which is really well described in their book uh I encourage you all to go and read through it it's fascinating uh so tokenization of intellectual property so ad can actually do some research and then distribute the outcomes of those research in form of intellectual property tokens uh which people can own trade exchange and so on and so on um out of that ecosystem uh there's a bunch of so-called bods okay and bods are small community research organizations focused on specific spe ific subjects right so we have signups da focused on brain health Vita da focused on longevity PID da focused on psychedelics uh Athena da Women Health valy da I don't remember um hair da that's an easy one so focused on hair cryo on cryo preservation and cerebrum Dow uh you may guess focused on brain health so all of those Dows raised millions of dollars um and they are funding research in specific areas of Interest uh there is a collaborative um community of scientists who can get the research funded start you know doing the research and ultimately um and ultimately uh share the commercial value of intellectual property that derives from that research one of the Prime examples is VA da focused on longevity the oldest and the biggest one of the research Dows um then if we go back to the this value chain there is the execution layer there's a bunch of projects here here for example Amino chain genomes Dow data L that's us here here I can't move I can't zoom in okay um anyway so for example genomes da is a DNA Vault for your genomic data um that promises uh first making genomic data available to researchers um and second monetizing that data on the site of the patients uh we as data L out of promotion here so we are you know a research company essentially uh we have um our own blockchain tools for recruiting patients to Scientific trials and managing of their informed consents uh which is pretty simple technology so we just keep track of all the consents and all the legal work connected with doing research on the blockchain transparently and the patient states in full control um we are doing commercial research as well for example we uh did um an analysis of the breast cancer uh patients Pathways in Poland for one of the pharmaceutical companies we are also doing the First Community funded um registry for a rare disease called asset singamas deficiency in Poland also uh working with a bunch of scientists and um and Industry companies um I think I don't have time to show you our Tech unless we have a minute left uh anyway the way we use blockchain is for storing an audit trail of all the consent operations so if we go back to the principles of decentralized science um we are about onchain record of uh of scientific studies um and then finally if your research is successful then you perhaps want to publish the results there are currently two platforms uh that are dealing with publishing of scientific results the first one uh and probably the biggest is called research Hub uh it's co-founded by Brian Armstrong whom you may know as the co-founder of coinbase um and this is essentially a social network and um uh a journal for scientists to publish their scientific research to be rewarded with their coin uh and kind of you know Distributing bounties working together um doing peer reviews for uh each other and uh creating this kind of you know boiling plate of scientific research and cooperation so think about it as a publishing paper and as a scientific social network with some uh incentive structure woven into it uh yeah so that's you know the example of research that they do um last but not least there's a fascinating project from the US and Switzerland called daps uh they are also a decentralized publishing Network so you can publish your results on the network create research objects uh that live on the blockchain you can publish your results but also your data sets so if somebody wants to replicate your study they can go over there you know read the whole logic read the whole paper download the data set do your own analysis um and be able to draw your own conclusions from that study right uh so that would be all I'm pretty much on time there's two minutes left if there are any questions there we go we can hear you all right right and that's the whole question so I think they didn't figure it out yet quite because right now you know the incentive is to publish in peer-reviewed journals and you know increase your age index and you know have citations and so on and based on that you climb the ladder of scientific ranks right now I think they don't have it figured good question I don't know so they are peer reviewed I think so uh they can put up you know a reward in the research coin for peer reviews I guess peers but who they are I'm not sure but that's a good question that's a good value we uh haven't used this platform yet uh because one of the research that I showed you on the previous slide we actually got it done just last week um but you know we are still considering whether it will be published on Research C perhaps yes for so in my basic understanding of the subject which may be wrong uh they have some form of you know legal mechanism of linking an nft token to the intellectual property from the research and then fractionalizing this nft with a bunch of utility erc20 tokens that are then distributed amongst the Dow members so that's my simple understanding but I may be wrong exactly exactly exactly that's all we have thank you