View Full Version : Genetics and heritability


Dizfriz
08-01-14, 10:49 AM
With all this discussion going on in regards to genes and ADHD I thought I might add in a little information.

Keep in mind that statistics of this type is not my field and it has been *many* moons since I had stat but I do have a basic groundwork which makes it a little easier for me to understand what is being said.

So if anyone is better trained in stat or has had more recent training please chime in to correct anything that I could have handled better.



Heritability means that the *variation* of a trait in a population (height or ADHD for example) is correlated with variation of the genetic make up of that population as well as environmental variation.

It is a fairly complex statistical construct and like correlation does not directly imply causation. In the case of ADHD, many studies indicate a heritability factor of around .8 which is a measure of the impact of the genetic make up of the population has on the variation or results so to speak.

This is not causation as statistics of this type do not get into causation but it is a measure of the influence genetics has on the disorder as compared with environmental factors.

So while it is not really inaccurate, in my opinion, to say that ADHD appears to be highly genetic in nature but it is more accurate, at least in a scientific sense, to say that ADHD is highly heritable which is a measurement of the role genes are playing in the incident rate of ADHD. Right now we cannot directly measure, for the most part, the impact of genes on ADHD but we can measure heritability which is a pretty good substitute.

In otherwords, we can safely say that ADHD is highly heritable with reasonable estimates of around .80 for heredity and around .20 for environmental factors involved in the variations of incident rates of ADHD in the population.


A couple of points


1. The number does not apply to individuals. It is a population measurement

2. The number is not fixed, differences in environmental factors can change the estimates of heritability.

As an example height. If you work with twin studies heritability of height is around .80, similar to ADHD. But (for example) if you compare height in one genetic population in an environment with limited food resources against that same population in an environment with plenty of food, you may get a situation where environmental factors play a larger role in the variation of heights. A good example is the Japanese in say the 1940's raised in Japan and Japanese raised in the U.S.. Differences in food availability correlates in a difference in population height.

Much of the work with this has been done with twin studies to help control for environmental variations such as this.


Hope this helps a little to bring some understanding into a complex subject. If you don't think it is complex read the Wiki article on heritability http://en.wikipedia.org/wiki/Heritability

Dizfriz

mildadhd
08-01-14, 12:24 PM
Lets say that 80% is heredity and 20% non-heredity

There are still environmental factors involved in the 80% heredity.

(specifics depend on family and individual circumstances)

Heredity is not only genetics, it involves epigenetics.

The brain develops in interaction with the environment.



P

Dizfriz
08-01-14, 12:42 PM
Lets say that 80% is heredity and 20% non-heredity

There are still environmental factors involved in the 80% heredity.

(specifics depend on family and individual circumstances)

Heredity is not only genetics, it involves epigenetics.

The brain develops in interaction with the environment.



P
1. Epigenetics does not = magic wand.

2. What does this have to do with a discussion on heritability as a statistical construct?

If you wish to discuss epigenetics perhaps you might take it to another thread. It is not relevant to this topic.


Dizfriz

mildadhd
08-01-14, 12:50 PM
Genetics and Heritability



We now have a general view of transcriptional regulation that can account for most observations that geneticists have made in the past century.

However, there are still some phenomena that beg for explanation.

An important set of phenomena, termed epigenetic inheritance ([URL="http://www.ncbi.nlm.nih.gov/books/NBK21276/)", seem to be due to heritable alterations in which the DNA sequence itself is unchanged.

Indeed, it is likely that these phenomena constitute another, poorly understood level of gene control.

Examples of epigenetic inheritance in which the activity state of a gene depends on its genealogical history are paramutation and parental imprinting.



P

mildadhd
08-01-14, 01:00 PM
"Also see heredity" (http://en.wikipedia.org/wiki/Heritability)

Heredity (http://en.wikipedia.org/wiki/Heredity) is the passing of traits to offspring from its parents or ancestor.

This is the process by which an offspring cell or organism acquires or becomes predisposed to the characteristics of its parent cell or organism.

Through heredity, variations exhibited by individuals can accumulate and cause some species to evolve.

The study of heredity in biology is called genetics, which includes the field of epigenetics.





i!i

Dizfriz
08-01-14, 02:20 PM
Peripheral, I am well aware of epigenetics. The thing is that this topic was not on this. '

You rather successfully hijacked my attempt to explain epigenetics and you are trying to do the same for this one

I am trying to explain and discuss the statistical construct of herdibility as at applies to the genetics of ADHD, not epigenetics. Please don't hijack this one too.

Dizfriz

mildadhd
08-01-14, 08:55 PM
Dizfriz,

I was not trying to derail this thread or any other threads.

In think information I was posting was directly related to the topics, in this and other thread your are bringing up.

I'm sorry you don't think so.

You posted these threads in response to other peoples discussions.

That being said, to avoid conflict , I won't reply to your threads or posts or questions anymore.


Take Care

P

Amtram
08-02-14, 10:51 AM
There are different kinds of heritability, and a condition that is present at birth is less likely to be turned on or off by environment as a condition that develops over time. A condition with a complex polymorphism will show a lot more variation given the influence of time and certain environmental factors than one that relies on a single gene mutation or Single Nucleotide Polymorphism.

The recent study identifying 108 gene loci implicated in schizophrenia provides a good example of this, but my geneticist and epigeneticist and neuroscientist tweeps warn against drawing conclusions about other conditions besides autism (lots of crossover with similar autism studies is why) from this. The study demonstrated not only the genetic heritability of the condition, but also the variability in effect. Each of the 108 mutations represent a number of different possible symptoms, but until they tease out which ones or which combinations influence which symptoms, we're not going to know which ones are causative and which ones represent predispositions.

The causative ones are going to be like monogenic conditions, in that the symptom they create is present regardless of outside influences. Much like a straightforward genetic condition, these symptoms might be subject to treatment, but will not be curable or preventable.

The predispositions will be more like physical diseases of the heart, or like certain types of cancers. Some will be difficult to control or prevent, and some will be highly responsive to preventative measures.

We like to think that we can control our destinies by making lifestyle changes, but in some cases all we can do is mitigate risk or control conditions. If, for example, your grandfather died at 40 from heart disease, your father died at 60 of heart disease because he ate right and exercised, then you can realistically expect that if you eat right and exercise and take medications, you might live a good amount of time longer than both of them before you succumb to heart disease, if you've inherited the genes that increase your risk of heart disease.

Similarly, if you have neurological/behavioral symptoms that arise from a condition of the brain like schizophrenia, there may be things that you can do to reduce your risk or treat the symptoms, because the condition usually comes on in your teens or twenties - but that doesn't mean you can custom tailor your environment to prevent it.

So with autism or ADHD, which are present from birth, you don't have the same opportunity to "plan." What you can do, though, is identify the symptoms (which may become easier once their genetic bases are found) and mitigate the outside causes that may exacerbate those symptoms. Like treating Juvenile Diabetes, you avoid things that make the condition worse and take medications to prevent the symptoms that can't be helped by avoiding environmental triggers. Only much more complex.

Dizfriz
08-02-14, 12:52 PM
Amtram, good post specially the last paragraph.

Genetics is complicated and the problem some run into is to try to simplify too much or completely misunderstand it and as a result become rather confused about what genetics and herdibility is and the effects. We don't do anybody any good by putting out confused or incorrect information.

I was seeing a good bit of misunderstanding when I would state that the heritability of ADHD is around .80 in a number of studies and that is why I posted what I did.

Heritability as a statistical representation of genetics is rather complex and since it the term in used a lot on the forum, I hope it did it well enough for an interested person without training in statistics or genetics could understand some of the issues and how statistical heritability is used.

The more I got into it the harder it was to simplify the concept. I will cross my fingers, maybe the post will help.

Dizfriz