is historic Ne useful? (e.g. estimated using Fastsimcoal or dadi)

7 views
Skip to first unread message

Alicia Mastretta

unread,
Nov 23, 2022, 5:39:54 PM11/23/22
to Genetic indicators project
Hi,

For some of the species we are assessing there are Ne estimations based on the site frequency spectrum (SFS) using coalescence methods (Fastsimcoal) or diffusion approximation (dadi). These methods are not available in our list of methods in the kobo form.

Since these methods provide a historical or long term Ne instead of a more recent one, I suppose these Ne estimators would not be useful for our purposes. Is this correct or should we incorporate these Ne estimators?

I remember we discussed this at some point, but I can't find the notes (if they exist) or a mention in our manuals.

Cheers,

Alicia




Joachim Mergeay

unread,
Nov 23, 2022, 5:43:36 PM11/23/22
to genetic-indic...@googlegroups.com
Indeed not very useful here.
J

Op wo 23 nov. 2022 om 23:39 schreef Alicia Mastretta <tic...@gmail.com>
--
You received this message because you are subscribed to the Google Groups "Genetic indicators project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/090c6381-454a-4c35-a8b9-9688e3a6525fn%40googlegroups.com.
--
Joachim Mergeay (he/him)
Research Institute for Nature and Forest - Belgium                                  
Tel: +32 499 942 942
Meldpunt wolven Vlaamse overheid : wo...@inbo.be
///////////////////////////////////////////




Fumiko ISHIHAMA

unread,
Nov 23, 2022, 6:20:19 PM11/23/22
to genetic-indic...@googlegroups.com
Hi,

Isn't there a possibility of estimating current Ne using historical and current distribution area based on SDM by assuming a constant ratio of Ne : (distribution area)?

This assumption is supported by the "mutation-area" relationship
https://www.science.org/doi/pdf/10.1126/science.abn5642
(The paper probably Joachim told us)

If it is possible, it would be very useful for species/ecosystems like the tropical regions where it is very difficult to estimate Nc due to low population density.


Fumiko


________________________________________
差出人: 'Joachim Mergeay' via Genetic indicators project <genetic-indic...@googlegroups.com>
送信日時: 2022年11月24日 7:43
宛先: genetic-indic...@googlegroups.com
件名: Re: is historic Ne useful? (e.g. estimated using Fastsimcoal or dadi)

Indeed not very useful here.
J

Op wo 23 nov. 2022 om 23:39 schreef Alicia Mastretta <tic...@gmail.com<mailto:tic...@gmail.com>>
Hi,

For some of the species we are assessing there are Ne estimations based on the site frequency spectrum (SFS) using coalescence methods (Fastsimcoal) or diffusion approximation (dadi). These methods are not available in our list of methods in the kobo form.

Since these methods provide a historical or long term Ne instead of a more recent one, I suppose these Ne estimators would not be useful for our purposes. Is this correct or should we incorporate these Ne estimators?

I remember we discussed this at some point, but I can't find the notes (if they exist) or a mention in our manuals.

Cheers,

Alicia






--
You received this message because you are subscribed to the Google Groups "Genetic indicators project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-...@googlegroups.com>.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/090c6381-454a-4c35-a8b9-9688e3a6525fn%40googlegroups.com<https://groups.google.com/d/msgid/genetic-indicators-project/090c6381-454a-4c35-a8b9-9688e3a6525fn%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
Joachim Mergeay (he/him)
Research Institute for Nature and Forest<https://www.inbo.be/> - Belgium<https://www.inbo.be/>
Associate Professor KULeuven Ecology Evolution and Biodiversity Conservation
<https://bio.kuleuven.be/eeb>
Tel: +32 499 942 942
Meldpunt wolven Vlaamse overheid : wo...@inbo.be<mailto:wo...@inbo.be>
///////////////////////////////////////////
[https://inbo-website-prd-532750756126.s3-eu-west-1.amazonaws.com/inbologoleeuw_nl.png]




--
You received this message because you are subscribed to the Google Groups "Genetic indicators project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-...@googlegroups.com>.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/CAFP9MUu4iJw6z9Xw5r092P4DzWM1F6%3DoBDjVNb3z6sxr9zAK8g%40mail.gmail.com<https://groups.google.com/d/msgid/genetic-indicators-project/CAFP9MUu4iJw6z9Xw5r092P4DzWM1F6%3DoBDjVNb3z6sxr9zAK8g%40mail.gmail.com?utm_medium=email&utm_source=footer>.

Myriam Heuertz

unread,
Nov 24, 2022, 4:54:11 AM11/24/22
to genetic-indic...@googlegroups.com
Hi, no trying to post from my other email address!
Cheers,
M

On Thu, 24 Nov 2022 at 10:48, Myriam Heuertz <myriam....@inrae.fr> wrote:
Hi Alicia, hi all,

I have been thinking of this as well. I think that there is consensus that historical Ne (ancestral population size) cannot substitute a contemporary Ne measurement for conservation management. 
However, assuming that estimates of historical and contemporary Ne are comparable (this is not trivial, due to sampling and methodological aspects), the comparison can inform on the genetic load a population currently carries. Historically small populations will have purged much of their genetic load, whereas historically large populations will retain many deleterious mutations that can become unmasked when population size (Ne) crashes. Therefore populations that were ancestrally large are expected to be more genetically vulnerable if they experience a current low Ne. From a population genetics point of view, to estimate population extinction risk, historical Ne thus carries additional information that contemporary Ne capture. However if we included ancestral Ne in our Kobo form (some standardization in terms of numbers of generations in the past may be needed), the risk is that both Ne estimates may be confounded by colleagues filling the Kobo form, and we want to make sure that we control this risk.

For further discussion in the next meeting, I'd say. I include this as an agenda item!

All the best,


To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/TYAPR01MB24795393644A9E7D642CA4ABE70C9%40TYAPR01MB2479.jpnprd01.prod.outlook.com.


--

Myriam Heuertz (DR2)
Biogeco, INRAE, Univ. Bordeaux

Mailing address:
INRAE UMR BIOGECO 1202
Domaine de l'Hermitage
69 route d'Arcachon, CS 80227
33612 Cestas Cedex, France

Phone: +33 5 35 38 5329 (office), +33 780 41 33 85 (mobile)




--

Myriam Heuertz (DR2)
Biogeco, INRAE, Univ. Bordeaux

Mailing address:
INRAE UMR BIOGECO 1202
Domaine de l'Hermitage
69 route d'Arcachon, CS 80227
33612 Cestas Cedex, France

Phone: +33 5 35 38 5329 (office), +33 780 41 33 85 (mobile)


Joachim Mergeay

unread,
Nov 24, 2022, 5:17:44 AM11/24/22
to genetic-indic...@googlegroups.com
Hi Fumiko,

We use a simplified version of this in cases where we don't have Nc estimates. 
For example, take this case of the smooth snake in Belgium, Coronella austriaca.

In the literature we see that in optimal habitat densities are circa 1-2 per ha, and a dispersal capacity of circa 1 km. 
So we have taken all observations of this species for the past 10 years to define the occurrence, we plotted a 1 km radius on each observation, and all observations that are adjacent with this radius are considered part of the same (meta)population. 
In some areas we see non-adjacent observations despite continuous landscape features, so there also we allow to merge subpopulations together. 
Next, we mark a polygon contour along all supposed metapopulations and calculate the expected Nc (if all space in the polygon is suitable and occupied): so expected Nc= area x density. 
If that expected Nc < 5000 we consider this to be too small. if it is >5000 it is potentially large enough. 

Hope this helps

Joachim
Joachim 


On Thu, 24 Nov 2022 at 00:20, Fumiko ISHIHAMA <ishi...@nies.go.jp> wrote:
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/TYAPR01MB24795393644A9E7D642CA4ABE70C9%40TYAPR01MB2479.jpnprd01.prod.outlook.com.


--
Joachim Mergeay (he/him)
Research Institute for Nature and Forest - Belgium                                  
Tel: +32 499 942 942
Meldpunt wolven Vlaamse overheid : wo...@inbo.be
///////////////////////////////////////////




Fumiko ISHIHAMA

unread,
Nov 25, 2022, 2:17:02 AM11/25/22
to genetic-indic...@googlegroups.com
Hi Joachim,

Thank you for sharing your example and the beautiful map.
Such a method would be very useful if sufficient observation records and information about density and dispersal exist.

Thank you,

Fumiko


________________________________________
差出人: 'Joachim Mergeay' via Genetic indicators project <genetic-indic...@googlegroups.com>
送信日時: 2022年11月24日 19:16
宛先: genetic-indic...@googlegroups.com
件名: Re: is historic Ne useful? (e.g. estimated using Fastsimcoal or dadi)

Hi Fumiko,

We use a simplified version of this in cases where we don't have Nc estimates.
For example, take this case of the smooth snake in Belgium, Coronella austriaca<https://drive.google.com/file/d/1Dk4mBdUkOMsy96pnOBAz49PGar5nesCP/view?usp=share_link>.

In the literature we see that in optimal habitat densities are circa 1-2 per ha, and a dispersal capacity of circa 1 km.
So we have taken all observations of this species for the past 10 years to define the occurrence, we plotted a 1 km radius on each observation, and all observations that are adjacent with this radius are considered part of the same (meta)population.
In some areas we see non-adjacent observations despite continuous landscape features, so there also we allow to merge subpopulations together.
Next, we mark a polygon contour along all supposed metapopulations and calculate the expected Nc (if all space in the polygon is suitable and occupied): so expected Nc= area x density.
If that expected Nc < 5000 we consider this to be too small. if it is >5000 it is potentially large enough.

Hope this helps

Joachim
Joachim


On Thu, 24 Nov 2022 at 00:20, Fumiko ISHIHAMA <ishi...@nies.go.jp<mailto:ishi...@nies.go.jp>> wrote:
Hi,

Isn't there a possibility of estimating current Ne using historical and current distribution area based on SDM by assuming a constant ratio of Ne : (distribution area)?

This assumption is supported by the "mutation-area" relationship
https://www.science.org/doi/pdf/10.1126/science.abn5642
(The paper probably Joachim told us)

If it is possible, it would be very useful for species/ecosystems like the tropical regions where it is very difficult to estimate Nc due to low population density.


Fumiko


________________________________________
差出人: 'Joachim Mergeay' via Genetic indicators project <genetic-indic...@googlegroups.com<mailto:genetic-indic...@googlegroups.com>>
送信日時: 2022年11月24日 7:43
宛先: genetic-indic...@googlegroups.com<mailto:genetic-indic...@googlegroups.com>
件名: Re: is historic Ne useful? (e.g. estimated using Fastsimcoal or dadi)

Indeed not very useful here.
J

Op wo 23 nov. 2022 om 23:39 schreef Alicia Mastretta <tic...@gmail.com<mailto:tic...@gmail.com><mailto:tic...@gmail.com<mailto:tic...@gmail.com>>>
Hi,

For some of the species we are assessing there are Ne estimations based on the site frequency spectrum (SFS) using coalescence methods (Fastsimcoal) or diffusion approximation (dadi). These methods are not available in our list of methods in the kobo form.

Since these methods provide a historical or long term Ne instead of a more recent one, I suppose these Ne estimators would not be useful for our purposes. Is this correct or should we incorporate these Ne estimators?

I remember we discussed this at some point, but I can't find the notes (if they exist) or a mention in our manuals.

Cheers,

Alicia






--
You received this message because you are subscribed to the Google Groups "Genetic indicators project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-project%2Bunsu...@googlegroups.com><mailto:genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-project%2Bunsu...@googlegroups.com>>.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/090c6381-454a-4c35-a8b9-9688e3a6525fn%40googlegroups.com<https://groups.google.com/d/msgid/genetic-indicators-project/090c6381-454a-4c35-a8b9-9688e3a6525fn%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
Joachim Mergeay (he/him)
Research Institute for Nature and Forest<https://www.inbo.be/> - Belgium<https://www.inbo.be/>
Associate Professor KULeuven Ecology Evolution and Biodiversity Conservation
<https://bio.kuleuven.be/eeb>
Tel: +32 499 942 942
Meldpunt wolven Vlaamse overheid : wo...@inbo.be<mailto:wo...@inbo.be><mailto:wo...@inbo.be<mailto:wo...@inbo.be>>
///////////////////////////////////////////
[https://inbo-website-prd-532750756126.s3-eu-west-1.amazonaws.com/inbologoleeuw_nl.png]




--
You received this message because you are subscribed to the Google Groups "Genetic indicators project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-project%2Bunsu...@googlegroups.com><mailto:genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-project%2Bunsu...@googlegroups.com>>.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-project%2Bunsu...@googlegroups.com>.
Research Institute for Nature and Forest<https://www.inbo.be/> - Belgium<https://www.inbo.be/>
Associate Professor KULeuven Ecology Evolution and Biodiversity Conservation
<https://bio.kuleuven.be/eeb>
Tel: +32 499 942 942
Meldpunt wolven Vlaamse overheid : wo...@inbo.be<mailto:wo...@inbo.be>
///////////////////////////////////////////
[https://inbo-website-prd-532750756126.s3-eu-west-1.amazonaws.com/inbologoleeuw_nl.png]




--
You received this message because you are subscribed to the Google Groups "Genetic indicators project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to genetic-indicators-...@googlegroups.com<mailto:genetic-indicators-...@googlegroups.com>.
To view this discussion on the web, visit https://groups.google.com/d/msgid/genetic-indicators-project/CAFP9MUvWUo0ihFYvCm-WUhcFYM9zg01pRfOWxBedXXq1sv6n0Q%40mail.gmail.com<https://groups.google.com/d/msgid/genetic-indicators-project/CAFP9MUvWUo0ihFYvCm-WUhcFYM9zg01pRfOWxBedXXq1sv6n0Q%40mail.gmail.com?utm_medium=email&utm_source=footer>.
Reply all
Reply to author
Forward
0 new messages