Richard Graves (00:01.013)
Well Marlon, I'm delighted to welcome you on this week's edition of the Science for Sport podcast. Welcome to the show.

Marlon (00:07.576)
Thank you for having me.

Richard Graves (00:09.621)
Now let let's talk about what what you're doing with your company Bial is because fundamentally it has the potential to to be a game changer in terms of athlete testing. And yet, as far as I'm aware, your background i is a little bit unusual because you didn't follow the usual sports science pathway. Your background is is based in chemistry. So tell us a little bit about that and how you go from being a chemist in essence to having an interest in measuring.

what's happening inside an athlete through testing their saliva.

Marlon (00:43.534)
Yeah, thank you. Yeah, that's absolutely true. I am a chemist by training. I also worked in experimental endocrinology during my chemistry bachelor's where I did non-invasive steroid testing for animals. So this was basically the first touch points which I had with this kind of topic. So you could say it was a pretty shitty job because we did everything except blood. So it was mostly feces, hair and saliva.

Mostly stress testing. But then I went into medicinal chemistry. So basically organic chemistry, chem informatics, whole medicinal chemistry bang. And then during the PhD, COVID came along. And the lab was locked down. And I would say that's quite a bummer as a lab rat not being able to go to the lab. So for, okay, maybe instead of building molecules, I can build a company.

And you know, the whole world had some touch points now with saliva testing, although in a very different setting. But then this experience from the veterinary university came back to me and you know, one thing led to another. So it went from animals to humans to professional athletes. And to be honest, like I said, I'm not going from sports science and to be frank, I also was not too much interested in football or in elite sports in general. So for me, this was more like a natural progression.

Richard Graves (02:11.147)
Why saliva then? What what what was the decision behind that? What does saliva tell us and what properties does it contain that can give us a a greater understanding of what's happening inside an athlete's body?

Marlon (02:23.074)
Yeah. So maybe we need to understand what saliva is or how we can frame it. saliva is sometimes also described as being like an ultra filtrate of blood. So it's basically filtered blood, which leads to saliva having a big overlap in biomarkers compared to blood. But on the other hand, it's very challenging matrix. So the concentrations in saliva are very low. So this is also why it's so hard to analyze it. But at the end of the day,

It's the only matrix that's really accessible if you want to test in high frequency and this is what we're doing. So for us, it's not about one biomarker, it's about many biomarkers describing potential internal load like a multi-organ system. So we have biomarkers covering sleep, psychological stress, immunological biomarkers, inflammation, hydration, all kinds of different things. So we try to make sense out of it and we need to do this in a high frequency setting. So testing.

as much as possible, as often as possible, but in a high depth. So you can really understand what's going on, trends, patterns, things like this. So you're not even competing with point of care products at this stage because there's usually about like one or two biomarkers which you can access like in a very short period of time. I mean, we're still fairly fast with the depth we are providing, but for us it's more about trends and chronics issue we try to uncover.

Richard Graves (03:48.96)
Now traditionally of course blood testing has been the way that sports science has measured these these biomarkers. Does saliva in itself tell us more? Does it tell us different things, different information, or do you see it as sitting nicely alongside blood testing?

Marlon (04:08.246)
It really depends on the biomarkers, to be honest, right? So sometimes it can tell you the same thing. Sometimes it's a different thing. I give you a very nice example here. So hormones, right? So if you assess hormones in blood, you usually get a total hormone level back. So let's say total testosterone levels. So all the testosterone in your blood, which is also bound to proteins, for example.

But when we test hormones in saliva, we only access the bioavailable part of the hormones, which is theoretically a bit more useful because you can basically see what's actually active in your body from a concentration perspective when it comes to hormones. So in this case, it's kind of similar, but it tells a bit different story. Although you can access this also for blood, but it's a lot more expensive and tedious to do it. So it really depends.

Richard Graves (05:05.717)
It goes out saying I don't have your your wealth of of knowledge or or insight when it it comes to this level of of science. But I will tell you from reading the literature and listening to you already mentioning that word several times in this conversation. It's the frequency of testing that grabbed my attention. I'm sure it's the same for our listeners and viewers across the Science for Sport YouTube channel. explain why increased frequency of testing is important and

Potentially, does it enable us to form a a better picture of what's happening?

Marlon (05:41.004)
Yeah, absolutely. I mean, again, right. Blood is a super informative matrix and you can get a lot out of it, but it's not very accessible because I mean, even the biggest teams in the world when it comes to Venice in-depth blood tests and to do it maybe four times a season if they're hardcore, maybe eight times, something like this. Aside from that, they do some pinprick testing and having the possibility to basically test in depth on a daily basis because it's just so easy to do.

you can get so much more out of it. So we can build like individual baselines a lot easier than it would be possible with blood. So something we can assess in weeks would take years to be generated in the blood setting. So you really start to understand what's going on and figure out trends and patterns and systematic issues. Otherwise you would not be able to do in a blood setting.

Richard Graves (06:36.885)
And this is where it becomes particularly relevant for elite level athletes, sports sports science departments, organizations and teams, I imagine, isn't it? Because up until now, you you do the blood tests, like you say, six, seven, eight times a year, and it's compared to what is the given norm, the average of any any size of group, if you like. If you're able to test on a much more frequent basis, suddenly that baseline changes, doesn't it? And we start looking

Marlon (06:54.667)
Yeah. Yeah.

Richard Graves (07:06.103)
and begin to understand what normal looks like for an individual.

Marlon (07:09.357)
Yeah, 100%, 100%. So that's a very important point, right? So when you get your blood results and you're in the green, you're in the green with 95 % of the whole population, which also for professional athlete makes no sense at all, basically. So it's also not like designed for professional athletes. It's like for a broad population. So this changes quickly when you do like this high frequency saliva testing. And yeah, I mean...

It's also adaptive. also what we usually do is not only have like reference ranges, but like adaptive reference ranges with different sensitivities. So you can see how it changes during the season. I can tell like a pre-season testosterone levels can look very different than like in season or end of the season and things like this. So you can see how the biochemistry also changes over the weeks. So this is also really cool.

Richard Graves (08:00.823)
I know you're already working with individuals and organisations in the Bundesliga, in in the Premier League in the UK, you're working with elite cyclists across Europe as well. To what extent does this have the potential then to to be a game changer when we're talking specifically about elite sports, elite level athletes?

Marlon (08:22.079)
Yeah, so I mean, we are already at a stage where we can uncover problems which are not otherwise accessible to figure them out. So would say in general, we are very good at figuring out unknowns or hidden problems. So in that regard, we already doing a lot of unique things. But on the long term, it will be a lot also about injury risk predictions and things like this where you also need a lot of data. So this is also something we're working on because I mean,

Even if I say I have like 30 injuries in a season or whatever, it's like 15 are non-contact, 15 are contact, I'm interested in non-contact injuries, then you drill it down at the end, you landed like three or four injuries, which is very difficult to build any robust model, which changes if you have a lot of teams, a lot of data. And this is also currently where we're heading when it comes to like this holy grail goal of the company, when it comes to like injury projections and things like that.

Richard Graves (09:17.673)
Of course, there is always a caveat to things and topics like this. You're producing this wealth of data, there's frequency of testing, you have great analysis of of biomarkers within an individual. But as we always say on this podcast, context is key to anything that that is done. So how do you turn all this biomarker data into a format that sports scientists can act upon and coaches can utilize to a constructive capacity in training schedules and organisations?

Marlon (09:32.043)
How's it?

Marlon (09:37.517)
Yeah.

Marlon (09:47.692)
Yeah, that's a very, very big topic for us. And like I said, right, we are generating then potentially a hundred of biomarkers for dozens of athletes every day. It's basically impossible to crunch for all this kind of data as a human. It's just too tedious. So this is something we work a lot on. use a lot of different AI tools, build our own tools. So for us, AI is just another tool in the toolbox really than anything else. It's a deep tech company.

And so we try to aggregate this kind of data and compress it into something usable and bring it into the context. So you usually also get GPS data, wellness data, performance data, things like this, because at the end of the day, it's super crucial. I'll you very powerful, easy example. If you have a hard training or whatever and everything goes up, so what? You don't need us for that. So it's kind of obvious that your biomarkers will respond to a hard load.

but it gets interesting if you didn't do much and all your inflammation biomarkers go up and then it gets juicy, right? So these are the things where we kind of need to bring the context in to make sense out of it. So you can also digest this information more easily.

Richard Graves (11:01.013)
You said a little bit earlier, Marlon, that one of the things saliva testing allows us to do is measure things like fatigue, wellness, where where an athlete's body is at. Obviously cortisol i is one of the key biomarkers i associated with this and that dictates these levels. So what can testing for cortisol tell us that's happening within an athlete's body? And perhaps just as importantly, what can't it tell us?

Marlon (11:23.277)
Yeah, so cortisol actually is a very, very cool biomarker. We obviously also do cortisol and cortisone. The thing that cortisol tells you is that something is going on, not necessarily what is going on. And this is where we start to combine those kind of signals with other biomarkers, give you another example. Let's say we have an increase. So we've seen this also nowadays. We have an increase in cortisol over several days and then the athlete gets sick, right?

then you have like an increased cortisol level for several days. And then it turns out this was a stress event. The trainer was sacked. look very similar for a lot of athletes. Or you have one isolated cortisol peak, which is extremely high, but it's just this one. And then it turns out the athlete didn't sleep too well. And now you start combining those signals, like the first signal, this increase with immunologic, your inflammation biomarkers. Okay. Maybe it goes more into like a.

direction that the athlete will get ill or something like that. Or you combine the stress event or the cortisol levels with biomarkers, which are more sensitive to psychological stress like alpha amylase from saliva and say, okay, maybe more, it's more a problem of psychological level. Or you combine this increase in cortisol, this isolated cortisol spike with melatonin levels and say, okay, maybe it was a problem with his sleep. And then you start building. So we call it basically indicator.

You basically start building those kind of indicators which can tell you a bit easier what's going on without going through dozens and hundreds of biomarkers basically.

Richard Graves (12:59.335)
You I'm I'm pleased you brought up this point 'cause I think it's particularly salient. How with the increase increased frequency of testing, obviously you get repetitive data coming through and little fluctuations can be seen immediately, pretty quickly. How wary do we need to be of misinterpreting that data? Say there's one elevated reading. do you need to guard against being overreactive to to one particular reading?

Marlon (13:24.044)
Yeah, absolutely. And I'm also telling all our partners we work with that, know, one outliers, no outlier at the end of the day is about trends. it going up for a longer period of time? Is it happening at certain events? It's those are the things we want to basically work on to, to, to, help the staff also get better, just the athletes in, what they do and not to be too reactive about it. So, so this is why.

It's also not so important for us to have the result in one or two hours like you do with a point of care system. It's more like this big picture stuff we are trying to assess here.

Richard Graves (13:58.976)
Now I imagine on on the flip side, the positive of high frequency testing is one of the ambitions surely has to be the acceleration of pro prognostics. in a sports science context, explain to to our listeners exactly what we're talking about when we say prognostics.

Marlon (14:16.426)
Yeah, when it comes to prognostics. it's more about being able to get ahead of the problem. This is how I would summarize it. Not just telling you, okay, this is now the problem, but this will be a problem in the future based on what we see at B-trans pattern and whatever it is, right? So this is how I would generally summarize it. Be ahead.

Richard Graves (14:42.759)
Are we talking about a scenario where potentially we can get to a point where we're almost predicting fatigue, illness, under recovery, these kind of things that are absolutely vital at an elite level?

Marlon (14:55.753)
Yeah, absolutely.

Richard Graves (14:58.699)
Let's throw things forward ten years then, because th this in itself, as I say, is potentially a game changer i in itself, w in a time when we we live in a society where everything wants to be instant, needs to happen quicker, don't don't we? In in your opinion, where do you think this could take us a decade from?

Marlon (15:17.42)
So from a decade from now, mean, the company has also big vision when it comes to the platform we're building, which goes also way beyond sports. So I think it's a great incubator and we will be helping a lot of athletes to be the best version of themselves basically. But in the long term, we want to make the platform also accessible for basically everyone in the world. So this is where we're heading long term when it comes to...

vision of Biolist, democratizing saliva biomarker testing. when it comes to being able to do predictions, improve the prognostic power of the platform, it will not take 10 years for elite sports. This will be a lot quicker because the data we are collecting is so dense, so rich, and all the context we get is just amazing you would never get in any other field, which also makes professional sports so compelling for us as a company.

especially as a deep tech company. But the long-term goal is really to bring this technology to the masses and be able to help them get, improve their health basically.

Richard Graves (16:29.043)
A and if that progresses i in a fashion which you envisage, Marlon, we we live in a technological age, don't we? What what is the percentage i in your mind that's person human led as opposed to AI led?

Marlon (16:46.73)
I mean, AI at end of the day is something we need to leverage to make it scalable. Because again, right? We talked about it's very complex data. have a lot of different things coming together. I mean, it's great as a human to kind of interpretate this and communicate the results and help the people. But if you really want to bring this to many people, there is another way around it to kind of leverage AI to make it like an automated process.

Richard Graves (17:17.399)
And right here, right now, I mean you've t told us the story. This basically came out of COVID as an idea and we're only five years on, so the growth has been remarkable. in your opinion, right here, right now, what's the the piece of the jigsaw that's missing and what's the next step?

Marlon (17:25.228)
That's right.

Marlon (17:37.696)
So I think the only bottleneck we have is time really. So it's like this exponential growth of data we are experiencing right now. So it took a lot of time to get started, to build some trust, to figure this out how we can actually do this in professional sports and create some value, but it gets faster and faster now. So I think it's just a matter of time.

This is our one.

Richard Graves (18:07.383)
Well, li listen, this has been a fascinating conversation. I feel enlightened by it because it's a topic that I was pretty naive towards before we started. So it's been fantastic having you on this week's podcast and I wish you all the best going forward. Before we let you go, for for those listeners like me that are hooked, where can they find you on social media and follow along the story?

Marlon (18:29.804)
Yeah, they can of course follow me on LinkedIn. I'm also posting a lot of stuff that's going on with our company. So please follow me on LinkedIn and of course also visit our website, Bioless.com.

Richard Graves (18:42.613)
Absolutely will do. All the best, Marlon.

Marlon (18:44.852)
It was great being here, Richard. Thank you.