UKBouldering.com

Genetic testing for performance (Read 11313 times)

bigironhorse

Offline
  • ****
  • junky
  • Posts: 767
  • Karma: +16/-0
    • YouTube
#25 Re: Genetic testing for performance
November 08, 2017, 11:44:57 am
Hmm.. I wonder if your relative lack of power is something to do with your genes. And if it is, whether there's a reliable test for this? :-\
 :tumble:
You won't ever get a sufficiently large sample size of climbers in which to conduct such a study specific to climbing.

I would 100% agree with this. GWAS style studies are notoriously low powered and its hard enough to pull out significant genes/variants even when you have a very large population of people with a strongly defined phenotype.

slackline

Offline
  • *****
  • forum hero
  • Posts: 18863
  • Karma: +633/-26
    • Sheffield Boulder
#26 Re: Genetic testing for performance
November 08, 2017, 12:02:40 pm
To clarify bigironhorse is referring to statistical power and not physiological power.

petejh

Offline
  • *****
  • forum hero
  • Posts: 5785
  • Karma: +623/-36
#27 Re: Genetic testing for performance
November 08, 2017, 01:14:34 pm
No there isn't a reliable test, and there won't be for a very long time.  People don't even know the complete set of genes involved, let alone quantifying all of the variation that exists within them throughout the human population which would then allow such tests to be performed.

The BioBank paper above is how to go about tackling the first step.  You won't ever get a sufficiently large sample size of climbers in which to conduct such a study specific to climbing.

Just for the record I 'was' being sarcastic..


I started this thread because it's an interesting topic and I'm curious to know more, so thanks Slackers for educating us all!

Correct me if I'm wrong as I'm sure you will but it seems to me that when we start talking about a lot of things to do with the human body/health/fitness/nutrition, it'll always be a very long time for scientists to be able to say with certainty that x+y = z. The best you can ever really hope for in some cases are strong correlations.

So saying something is or isn't scientifically proven doesn't appear to me to necessarily be a realistic standard to expect to achieve when it comes to talking about subjects like fitness, training, nutrition etc.

We follow training and nutrition protocols all the time that aren't scientifically 'proven' - some of the better training ideas out there no doubt based on scientific method, is backed up by previous training methodologies that have seemed to work and are based on established physiological research, and have a bunch of data from other climbers showing correlation with 'improvement x'. But I doubt that any of it is 'proven' to the scientific standards that are being discussed here.

I think it's normal for people to be attracted to the idea of finding out if they posses 'xyz' (or GCTA..) which in some tests has shown to be correlated to 'outcome abc'. Little harm in it as long as you possess healthy skepticism and a clear understanding of the limitations.

Doubt I'll be paying just yet though...


Oldmanmatt

Online
  • *****
  • forum hero
  • At this rate, I probably won’t last the week.
  • Posts: 7105
  • Karma: +368/-17
  • Largely broken. Obsolete spares and scrap only.
    • The Boulder Bunker climbing centre
#28 Re: Genetic testing for performance
November 08, 2017, 01:46:34 pm
Hmm.. I wonder if your relative lack of power is something to do with your genes. And if it is, whether there's a reliable test for this? :-\
 :tumble:
He doesn’t usually wear Jeans whilst Campusing, but he does have tendency to turn every session into an endurance fest. If only I could swap some of his endurance for a little of my strength; he’d be bounding up 8b+ in no time and I might manage to get beyond warm up...
I need to pinch a bit (actually a metric shit tonne) of foot work from him too.

slackline

Offline
  • *****
  • forum hero
  • Posts: 18863
  • Karma: +633/-26
    • Sheffield Boulder
#29 Re: Genetic testing for performance
November 08, 2017, 02:45:50 pm
Just for the record I 'was' being sarcastic..

I took the scratching of the chin to mean you were still wondering if there was such a test.



Correct me if I'm wrong as I'm sure you will but it seems to me that when we start talking about a lot of things to do with the human body/health/fitness/nutrition, it'll always be a very long time for scientists to be able to say with certainty that x+y = z. The best you can ever really hope for in some cases are strong correlations.

That is very much the nature of science, you have paradigms/models for viewing the world around you and go out and test the predictions based on that model, if the results match the predictions it lends support that the hypothesis is correct.  If evidence accumulates that the paradigm/model is wrong then the hypothesis is revised.

We follow training and nutrition protocols all the time that aren't scientifically 'proven' - some of the better training ideas out there no doubt based on scientific method, is backed up by previous training methodologies that have seemed to work and are based on established physiological research, and have a bunch of data from other climbers showing correlation with 'improvement x'. But I doubt that any of it is 'proven' to the scientific standards that are being discussed here.

I think it's normal for people to be attracted to the idea of finding out if they posses 'xyz' (or GCTA..) which in some tests has shown to be correlated to 'outcome abc'. Little harm in it as long as you possess healthy skepticism and a clear understanding of the limitations.

And therein lies the problem, the vast majority of people do not possess healthy skepticism nor do they think in complex terms because its just not at all simple in any way shape or form.  Hell many people still believe in deities and don't dare to question their beliefs?  There is actually huge scope for harm from unregulated genetic tests such as these, 23andme were prevented from providing any information on the association of genotypes with disease risk by the FDA in the US, although that has been relaxed a bit.  Why is it harmful?  Because its not a simple relationship and the vast majority of people don't understand the context of the information, nor will they be interested in learning it.

Most medical/nutritional/training research follows the research paradigm proposed by Rose in his book The strategy of preventive medicine where in essence you look at a sample from a population, if you can affect a change in the populations distribution, e.g. shifting the mean cholesterol levels (which are a predictor for risk of cardiovascular diseases) downwards, then everyone will be "better off".  The implicit assumption is that this change in the population parameter (the mean) is because everyone's reduced a little. This very often won't be the case though, out of your sample you may have some with very high cholesterol whose levels were dramatically reduced (this will also reduce the variance/standard deviation).  In essence and this is at the heart of your question, not everyone is the same, nor responds the same.  A paper from last year goes into this in greater detail... The idea of uniform change: is it time to revisit a central tenet of Rose's "Strategy of Preventive Medicine" (the full PDF is available from various sources including a certain illegal Russian site that has a mirror hosted in Brazil).  This has led in recent years to notions of "repsonders" v's "non-responders" and the panacea of Personalised Medicine

Back to your original question though...

Before asking what determinants might predict training response I'd say a step or two back needs to be taken and an estimate of the heritability, if any, of training response needs to be obtained.  This isn't straight-forward as off-spring share 50% of their genetics with each parent and they very often share the same environment, so you need to disentangle nature and nurture.  A classic approach to this in animals is selective breeding, not so popular in humans, so the alternative of Twin Studies are employed.  Thing is the money available for doing much of this research simply does not exist, big players are cancer and dementia research as the demographics of the populations shift as they age.

Speculate and cogitate to your hearts content and remain skeptical of anything you read, but ScienceTM won't have any climbing specific answers any time soon.  Tom and his Lattice pals are taking a fairly systematic approach to collating and analysing data so that might reveal some interesting avenues on the training front, but they won't obtain blood samples and consent from the hundreds of thousands of people required to perform Whole Genome Screen Associations.

If you are really, really interested in learning more about human genetics I can highly recommend Strachan and Read's Human Molecular Genetics, it was reading this course textbook for a module of the same name at undergraduate rather than attending the boring lectures that got me interested and set me on my path of employment.  Its very accessible and older versions (I still have a copy of the first edition) are still highly relevant and cheaper.


And as an aside here's a paper I came across on a tiny study on nutrition and recovery in climbers, but the sample size is so small that the conclusions are effectively useless...

Potter, Julia A and Fuller, Belinda (2014) The Effectiveness of Chocolate milk as a Post-Climbing Recovery Aid. Journal of Sports Medicine and Physical Fitness, epub. ISSN 0022-4707 (In Press)

Its great that someone has learnt how to do research for their Masters thesis but to suggests such tiny studies have the same credibility as larger more rigorous research is a discredit to ScienceTM and increasingly the validity of that hallowed status for understanding the world is being eroded, helped in no small part by people wanting incredibly complex topics boiled down to sound bites and headlines that give a binary answer.  Such papers should not be permitted to be published in peer-reviewed journals at all in my view.


abarro81

Online
  • *****
  • forum hero
  • Posts: 4303
  • Karma: +345/-25
#30 Re: Genetic testing for performance
November 08, 2017, 03:00:09 pm
Tom - why do you think excessive weight fluctuations limit gains? Per our conversation the other day, surely holding too low a weight through the year is more likely to inhibit gains? No doubt that holding holding a healthy, slightly higher weight the whole time is better for gains than fluctuations, but also worse for performance (links to the balance you mentioned obvs)

Murph

Offline
  • ****
  • forum abuser
  • Posts: 653
  • Karma: +66/-0
#31 Re: Genetic testing for performance
November 10, 2017, 02:46:34 pm
Barrows- In the absence of Tom's reasoning I'll offer my view on weight fluctuations being bad. Causative or not, they could well be indicative of a lack of consistency. And consistency, it has been said, is the key to success.

Even among people who employ some sort of bulk cut strategy they aren't going to be going up and down like a yoyo.

Your example of holding too low a weight, yeah that'd be bad too. But so too would holding much too high a weight!

Tommy

Offline
  • ****
  • junky
  • Posts: 814
  • Karma: +97/-1
#32 Re: Genetic testing for performance
November 10, 2017, 08:19:54 pm
Barrows- In the absence of Tom's reasoning I'll offer my view on weight fluctuations being bad. Causative or not, they could well be indicative of a lack of consistency. And consistency, it has been said, is the key to success.

Even among people who employ some sort of bulk cut strategy they aren't going to be going up and down like a yoyo.

Your example of holding too low a weight, yeah that'd be bad too. But so too would holding much too high a weight!

Hey Alex, sorry forgot to answer this. It's not that I think that weight fluctuation will definitely cause people to lack response to training (although there is still an element to this - just look at the effect on natural testosterone levels) but more that it can lead people to feel like they are "non-responders". I'll give you an example.

Climber X in 2016 weighs 70kg and climbs V10, but that year drops 2kg and bumps up to V11. He feels good about this & will often assign some of this gain to fantastic training methods in that same year, which is more or less not really true.

Climber X in 2017 goes back to 70kg during training season but struggles (for whatever reason) to drop to 68kg the following summer. They don't achieve V11 or V12 they were hoping for and get despondent. Typically (in my experience) a lot of climbers will then think their training is not effective rather than being objective about their weight. The fluctuating weight can create a lot of "noise" in performance levels and make it hard for the self-managed climber to maintain objectivity.

Thus, my thought is that people who go up and down quite a lot can erroneously assign performance gains (or losses) to their training and not be objective enough to understand the effect of weight. This becomes a long term habit and then eventually becomes deeply engrained in performance psychology.

In your case Alex, I don't think you'll see this as you're really disciplined, really honest with yourself and know why and when you're going to be light and heavy. 99% of people are not like this though!

Murph

Offline
  • ****
  • forum abuser
  • Posts: 653
  • Karma: +66/-0
#33 Re: Genetic testing for performance
November 10, 2017, 10:41:45 pm
Must admit when I saw "excessive weight fluctuations" I wasn't imagining +/-1kg... what's a V grade worth these days Tom? Just 2kgs or so? Incredible! :blink:

Fultonius

Offline
  • *****
  • forum hero
  • Posts: 4331
  • Karma: +138/-3
  • Was strong but crap, now weaker but better.
    • Photos
#34 Re: Genetic testing for performance
November 11, 2017, 08:37:49 am
Yeah, tom, can you clarify if you are talking about 1 or 2kg, or 3 to4 kg?  I fluctuate 1 kg weekly!



We follow training and nutrition protocols all the time that aren't scientifically 'proven' - some of the better training ideas out there no doubt based on scientific method, is backed up by previous training methodologies that have seemed to work and are based on established physiological research, and have a bunch of data from other climbers showing correlation with 'improvement x'. But I doubt that any of it is 'proven' to the scientific standards that are being discussed here.


I also wonder how much psychology plays its part - i.e. if someone comes up with the latest best training plan with top level climbers getting good results, is it necessarily better than what came before, or does it's newness and association just inspire people to train harder and be more focused, therefore getting better results?

Tommy

Offline
  • ****
  • junky
  • Posts: 814
  • Karma: +97/-1
#35 Re: Genetic testing for performance
November 11, 2017, 08:54:56 am
More clarifications :-)

Weight change = change in body fat, not "I'm dehydrated" / "ate a big curry" / "feel bloated". I think we all know that daily fluctuations mean very little, although the fact that this has been brought up suggests maybe not?!

True 1kg (up or down) = very little

2-4kg  (up or down) = significant (hence my example) but as I work as a coach I don't really want to create examples that make weight loss look too appealing e.g. saying client loses 8kg and boulder grade goes up 4 grades (which it most likely would if they weren't already lean).

+4kg (up or down) = excessive

Like ALL situations this is hugely dependent on the individual. On UKB you've got top end athletes like Barrows, Mason etc and then some weekend warriors and also a load of desk jockeys who are dads, like wine and still like climbing.... basically you're all very different beasts!

And of course I'm also talking of constant swings here as well. A loss of 4kg could be brilliant for one individual, but I would assert that any climber who's constantly swinging +/-4kg through the seasons will find it pretty hard to be objective about their response to training.



abarro81

Online
  • *****
  • forum hero
  • Posts: 4303
  • Karma: +345/-25
#36 Re: Genetic testing for performance
November 11, 2017, 12:11:51 pm
Got ya.

 

SimplePortal 2.3.7 © 2008-2024, SimplePortal