Obese gene

Help Support Steer Planet:

knabe

Well-known member
Joined
Feb 7, 2007
Messages
13,639
Location
Hollister, CA
HerefordGuy said:
When we do genome-wide analysis for marbling in Angus, we don't see any large effects, all we see is thousands of small effects. I disagree that there is low-hanging fruit, especially for marbling.


i don't mean low hanging fruit for marbling, sorry, i mean how fat is deposited, which can include marbling, but since it's overall percentage is low, to me, that could break either way.


i guess one thing i don't understand or forgot, is how thousands of small effects don't get lost due to error, especially if the training population is diverse and the pathways have little overlap between individuals. to me, this marbling thing seems to be left over pathway energy as opposed to any direct effect.


i forget the papers, but there is these little cells, and their number and distribution and distribution density may vary, but for some reason, they fill with fat, or lose fat or something, depending upon available energy and it's use, ie foraging in the wild versus statically in a feedlot like a glutton.
 

knabe

Well-known member
Joined
Feb 7, 2007
Messages
13,639
Location
Hollister, CA
HerefordGuy said:
When we do genome-wide analysis for marbling in Angus, we don't see any large effects, all we see is thousands of small effects. I disagree that there is low-hanging fruit, especially for marbling.


is it possible to discuss how the animals were chosen for the GWA and reservations you had with regard to their accuracy and any other considerations?


did you make pools of samples?
 

HerefordGuy

Well-known member
Joined
Apr 21, 2009
Messages
442
Location
Sturgeon, MO
How much of the phenotypic or genetic variance is this leptin test accounting for?
If this leptin variant was a causal mutation of large (or moderate) effect, the p-values would be very small and highly significant. It would also be replicated across independent experiments. In 2006, Taylor's group tried to replicate the leptin result, and it was not significant.

The only genes I know of that would be worth introgressing would be CAPN1 and CAST for tenderness, PLAG1 for growth, and DGAT for milk protein. For quantitative traits, large effect gene variants are the rare exception, not the rule. Evolution uses thousands of small effects to influence the genetic variation for complex and quantitative traits. This is why genetic prediction (EPDs and PTAs) and genomic predictions work. With EPDs we use pedigrees and progeny records to account for thousands of small effects. With genomic predictions we use the sum of thousands of DNA markers to account for thousands of small effects.
 

HerefordGuy

Well-known member
Joined
Apr 21, 2009
Messages
442
Location
Sturgeon, MO
knabe said:
HerefordGuy said:
When we do genome-wide analysis for marbling in Angus, we don't see any large effects, all we see is thousands of small effects. I disagree that there is low-hanging fruit, especially for marbling.


is it possible to discuss how the animals were chosen for the GWA and reservations you had with regard to their accuracy and any other considerations?


did you make pools of samples?
Here is the research paper.
http://www.biomedcentral.com/1471-2164/13/606 Plots are in the supplementary material.
We collected as many Angus AI sires as we could get our hands on. Most of these represent Angus sires with large numbers of progeny, but also includes unique populations like the Wye herd.
We weighted each animal's EPD by its accuracy, less accurate EPDs received less emphasis in the analysis. We removed parent average effects from the EPDs. So, we explicit modeled the inaccuracy and dependency between samples in this analysis (by using a mixed-model analysis).
Other papers and other groups are seeing the same results, thousands of loci of small effect. Human research also finds these same results. Evolutionary theory also says we should expect these results.
 

HerefordGuy

Well-known member
Joined
Apr 21, 2009
Messages
442
Location
Sturgeon, MO
knabe said:
HerefordGuy said:
When we do genome-wide analysis for marbling in Angus, we don't see any large effects, all we see is thousands of small effects. I disagree that there is low-hanging fruit, especially for marbling.


is it possible to discuss how the animals were chosen for the GWA and reservations you had with regard to their accuracy and any other considerations?


did you make pools of samples?
Animals were not pooled, each animal was individually genotyped.
 

librarian

Well-known member
Joined
Jul 26, 2013
Messages
1,629
Location
Knox County Nebraska
Sorry, I was being a smart alec about introgression, never a good idea.  There is a lot of philosophical territory here that maybe isn't quite appropriate for Steerplanet, but why not?  From my 5 minute study of this word that is new to me (but the idea seems to guide me), introgression can occur naturally, or can be accomplished through genetic engineering.  A smart alec might suggest that we splice an intention to evolve into the genetic technology so that the organism can short circuit our scientific notions of what is profitable.
I am always interested in what old time breeders "see" when they study cattle. The breeder knabe referred to early in this discussion figured this out through observation, I assume, not genomic analysis.  Even if you ask and even if you are told what bloodline carries a trait, you have to be able to see  what you are looking at.  I have a friend who thinks people don't spend enough time on observation any more, because the world has gotten so full of demands on our time.  I think there is truth in that. It's like when big pharma synthesizes a drug that grows wild in the Amazon, making the Amazon irrelevant.
So, knabe says
i guess one thing i don't understand or forgot, is how thousands of small effects don't get lost due to error, especially if the training population is diverse and the pathways have little overlap between individuals. to me, this marbling thing seems to be left over pathway energy as opposed to any direct effect.
i forget the papers, but there is these little cells, and their number and distribution and distribution density may vary, but for some reason, they fill with fat, or lose fat or something, depending upon available energy and it's use, ie foraging in the wild versus statically in a feedlot like a glutton.
be says


That part gets more to what I am pondering.  What would be the natural history of that left over pathway and what would have been the trade off.  It seems analogous to obesity in native human populations who have evolved on a low carbohydrate diet and then been switched over to white flour and sugar.  Big trade off in survivability for an increased ability to store fat.
Wikipedia tells me, "Introgression is an important source of genetic variation in natural populations and may contribute to adaptation and even adaptive radiation.[1"
Our attempts to reclaim the ability to fatten on grass alone across diverse environments seems to call for releasing the gene pool from streamlined combinations of ultimate fitness for the feedlot to get on with re-adaptive radiation.
and this from knabe
What I'm really after within any breed is a way to capture the difference in expression and allow producers to introgress the combinations they want, validate them and make breed crosses if they so desire and make combinations they want.


How would you make this capture? Would the introgression be natural or technological?
 

librarian

Well-known member
Joined
Jul 26, 2013
Messages
1,629
Location
Knox County Nebraska
My whole long and winding road with crossbreeding started with reading about the Nazi's (sorry, but it was them) attempt to breed back to the Aurochs by combining and recombining all the available cattle genetics and selecting for Aurochs type.  At that time I read that they actually accomplished this in only about 12 years, with Heck's cattle. Turns out is wasn't quite like that, but I figured if they can re-breed an Aurochs, I should be able to re-breed a Scotch Shorthorn. 
As usual, one has to be careful what they wish for.  I think these Scotch types may have the obesity gene.
Now I'm trying to figure out if that is a good or bad thing.
 

knabe

Well-known member
Joined
Feb 7, 2007
Messages
13,639
Location
Hollister, CA
It makes sense that evolution generally wouldn't have large effect genes for many pathways.
 

knabe

Well-known member
Joined
Feb 7, 2007
Messages
13,639
Location
Hollister, CA
Sorry I meant pooling after data collection.

I will read paper.

Was there any pair wise substitution on associations to eliminate candidates or how did you eliminate candidates if effect is so low and were there any interesting dependicies?

Probably my choice of the word pooling is technically incorrect.
 

librarian

Well-known member
Joined
Jul 26, 2013
Messages
1,629
Location
Knox County Nebraska
I thought this was the cool part of the conclusions:

"Finally, our results suggest that natural selection has also acted in this domesticated population to increase immunity and possibly to buffer the organism against the effects of inbreeding depression"
 

RyanChandler

Well-known member
Joined
Jul 6, 2011
Messages
3,457
Location
Pottsboro, TX
librarian said:
My whole long and winding road with crossbreeding started with reading about the Nazi's (sorry, but it was them)

There is much to learn from their approach.  If only the Nazi ideology had been applied to livestock and not humans. 

'Keep your blood pure,
It is not yours alone,
It comes from far away,
It flows into the distance
Laden with thousands of ancestors,
And it holds the entire future!
It is your eternal life'

-There is no substitute for observation-
 

knabe

Well-known member
Joined
Feb 7, 2007
Messages
13,639
Location
Hollister, CA
Nazi model was flawed due to anthropogenic projection.

Model required constant repetition of a lie, almost chant like.

 

librarian

Well-known member
Joined
Jul 26, 2013
Messages
1,629
Location
Knox County Nebraska
Do you think the tt cows would have more fat deposition in their udders? 
I have heard that one of the reasons Shorthorns fell out of favor was that they got hog fat in feedyards after WWII when they started really pushing corn.  Is that accurate?
 

librarian

Well-known member
Joined
Jul 26, 2013
Messages
1,629
Location
Knox County Nebraska
And here is a picture. The caption goes:
Lutz Heck (far left) and Hermann Göring (not surprisingly, on the far right) examining a relief map of the Białowieża Forest. The tiny figures represent characteristic game animals (elk, red deer, etc.); in the background a stuffed European bison looms. The horn sitting on the map is perhaps from a Heck aurochs. Image from Waidwerk der Welt, a catalogue published to accompany the 1937 International Hunting Exposition in Berlin.
 

Attachments

  • Bialowieza_map_NEW.jpg
    Bialowieza_map_NEW.jpg
    52.2 KB · Views: 285

HerefordGuy

Well-known member
Joined
Apr 21, 2009
Messages
442
Location
Sturgeon, MO
librarian said:
And here is a picture. The caption goes:
Lutz Heck (far left) and Hermann Göring (not surprisingly, on the far right) examining a relief map of the Białowieża Forest. The tiny figures represent characteristic game animals (elk, red deer, etc.); in the background a stuffed European bison looms. The horn sitting on the map is perhaps from a Heck aurochs. Image from Waidwerk der Welt, a catalogue published to accompany the 1937 International Hunting Exposition in Berlin.
Because so much variation was lost when cattle were domesticated, I don't think you can reconstruct aurochs from domestic cattle. You can make something that looks similar, but the DNA is not necessarily the same.
 

HerefordGuy

Well-known member
Joined
Apr 21, 2009
Messages
442
Location
Sturgeon, MO
knabe said:
i guess one thing i don't understand or forgot, is how thousands of small effects don't get lost due to error, especially if the training population is diverse and the pathways have little overlap between individuals. to me, this marbling thing seems to be left over pathway energy as opposed to any direct effect.
Are you talking metabolic pathways?

The larger the effect the more accurate the estimated effect size is. The smaller the effect, the less accurately its effect is estimated. But, when we do the analysis, even though the effects are small they cluster in the same gene networks. In my mind this highlights that individually each gene/variant (which is tagged by a marker) has little/no predictive ability. But, if we look at hundreds or thousands of effects we now have predictive ability.

The populations are not diverse, they usually contain individuals from a single population/breed.

Because we are dealing with small effects, we need large sample sizes. That is why we don't even attempt to create a genomic prediction unless we have over 1,000 samples. The larger the sample, the better the effect is estimated.



 

librarian

Well-known member
Joined
Jul 26, 2013
Messages
1,629
Location
Knox County Nebraska
after that detour, back to the selection pressure for fat deposition question

Biosci Biotechnol Biochem. 2000 Oct;64(10):2240-2. http://www.ncbi.nlm.nih.gov/pubmed/11129604
Differential response of obese gene expression from fasting in bovine adipose tissues.
Kim H1, Chi Y, Chung K, Kim K, Choi Y, Baik M.
Abstract
To understand the molecular mechanism for intramuscular fat deposition, the expression of the obese gene was examined in response to fasting. Food deprivation for 48 h induced a decrease in the level of obese mRNA in pooled adipose tissues (abdominal, perirenal, subcutaneous, intermuscular and intramuscular). The expression of obese mRNA was examined for individual adipose tissue from several fat depots. It was highly expressed in perirenal adipose tissue, but fasting did not affect its expression level in this tissue. Moderate levels were detected in subcutaneous and intermuscular adipose tissues, and a fasting-induced decrease in obese mRNA was apparent in these tissues. The expression level of the obese gene in intramuscular adipose tissue was very low and did not respond to fasting.

So, yes, it seems to be about those little cells filling up or emptying out depending upon the energy requirements.  But, rather than thinking about this in terms of gluttony and a fast track to marbling, I am trying to think of it in terms of reproductive fitness.  I think it was Mill Iron who said that cows are supposed to fluctuate in weight between winter and summer, and his made up for their weight loss over the winter by getting fat in the summer.  Seems natural enough.  I read in one of those old accounts of cattle in Scotland, that the best meat was the "new meat" laid on in the summer after the animal had starved over the winter.  The experts always tell us once you lose that fat, the meat will never marble, but I wonder.

But about fertility and milk production vs. fat deposition.  I was thinking about my Galloway cattle, who seem to have not much milk, but what they have seems to be "thick" and "high test".  The Galloways also store fat in the muscle instead of between the muscles or as back fat.  Maybe they can get it out faster that way.  These cattle are evolved to survive hard times in a cold wet climate with scant forage. They fatten easily.
Studying that, I came up with this about the Obese Gene and Milk Fat
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0063406
excerpts for those who don't like to plow through all that jargon:

Although a number of cellular, molecular and genetic studies have been performed with FTO, its functions and effects are far from being understood. Even less is known about RPGRIP1L. Considering that the FTO gene including its linked genomic neighborhood affects fat deposition in humans, the question arises, whether this gene region also affects the amount of fat delivered in milk during lactation. If the FTO locus does not only affect fat synthesis but is also involved in the regulation of energy balance, we would also expect additional effects on other milk components.

Our findings suggest that the FTO region not only regulates milk fat yield, but also the total energy content of milk. With regard to GWAS in humans, the FTO region has been repeatedly associated with body mass index and obesity. However, studies with lean mass have not been performed. To further test the pleiotropic effects of the FTO region, the analysis of traits characterizing body composition would be of interest. But body mass measurement of dairy cattle is not a matter of routine under production conditions.

The effects of the genetic variation in the FTO region accounted for about 1% of the corresponding traits variance in the analyzed cattle population. Even if the effect is small, it seems to be consistent across species and deserves more attention as a factor contributing to complex traits, which are expected to be formed by small effects of many loci [7].

In summary, our study in dairy cattle provides evidence that the obesity-associated FTO gene region accounts for variation in milk fat yield. For the first time, we show that the region does not only control fat but also protein yield and that both milk composition traits are regulated in the same direction. Therefore, we suggest that the FTO gene region controls the energy amount secreted during lactation. The position of the associated haplotype blocks and SNPs, their direction of effect and allele frequency distribution detected in our cattle study suggest that at least two causative variants account for differences between genotype classes. These mutations most likely underlie different selection pressure for production traits. In turn, this indicates different biological functions of the involved gene variants with respect to control and regulation of fat and protein metabolic pathways and in regard to maintaining energy homeostasis and controlling energy partitioning. Besides FTO, the neighboring upstream gene RPGRIP1L and the downstream non-coding genes U6ATAC and 5 S rRNA have functional relevance for milk fat and protein yield.
 

knabe

Well-known member
Joined
Feb 7, 2007
Messages
13,639
Location
Hollister, CA
HerefordGuy said:
Are you talking metabolic pathways? yes

The populations are not diverse, they usually contain individuals from a single population/breed.
the sources of contributions are not diverse within a population?
the locations of effects are not diverse?


if one looks across breeds, understanding the divergence past the markers (or the marker not even being present, rendering them with lower utility, what percent snp difference is there over the entire regions of interest? what is the distribution of snps past the markers? ie are they random, in upstream, downstream regions, within introns, within exons or both compared to upstream downstream regions or is it all just random, understanding random is a relative term and certain regions have different levels of conservation/pressure.
 

HerefordGuy

Well-known member
Joined
Apr 21, 2009
Messages
442
Location
Sturgeon, MO
knabe said:
if one looks across breeds, understanding the divergence past the markers (or the marker not even being present, rendering them with lower utility, what percent snp difference is there over the entire regions of interest? what is the distribution of snps past the markers? ie are they random, in upstream, downstream regions, within introns, within [/color]exons or both compared to upstream downstream regions or is it all just random, understanding random is a relative term and certain regions have different levels of conservation/pressure.
We haven't had the sequence data to look at this yet.

The region of interest is the entire genome. We don't pick out gene regions in a genomic prediction. We fit the entire genome.
 
Top