PDA

View Full Version : Create custom group averages



MonkeyDLuffy
07-03-2020, 07:30 AM
Hi, I was wondering if its possible to create a group average? I would like to run my group average but I couldn't find any options.

Thanks!

GenoPlot
07-03-2020, 05:17 PM
Hi, I was wondering if its possible to create a group average? I would like to run my group average but I couldn't find any options.

Thanks!

We had held off on that feature pending the release of the sub-grouping feature within a user's custom samples. However, if you would like the ability to view the averaged results of a cohort of samples, along with their individual results, that can be quickly enabled.

We can also add the auto creation of custom group averages, but that would probably be of limited value without sub-grouping as the custom samples belonging to any one user may be from varied ethnicities.

Would it be helpful to enhance the Ghost creation feature to allow for averaging? You could select as many samples as you wished, from your own library as well as the G25 dataset and average them out to create a custom group average.

MonkeyDLuffy
07-03-2020, 11:42 PM
We had held off on that feature pending the release of the sub-grouping feature within a user's custom samples. However, if you would like the ability to view the averaged results of a cohort of samples, along with their individual results, that can be quickly enabled.

We can also add the auto creation of custom group averages, but that would probably be of limited value without sub-grouping as the custom samples belonging to any one user may be from varied ethnicities.

Would it be helpful to enhance the Ghost creation feature to allow for averaging? You could select as many samples as you wished, from your own library as well as the G25 dataset and average them out to create a custom group average.

Yes that's something should work. But I'm not sure how to do that, is there any tutorial for that? Thank you for responding.

The tool is really good, I love the PCA

GenoPlot
07-04-2020, 02:29 AM
Yes that's something should work. But I'm not sure how to do that, is there any tutorial for that?


We're getting ready to release some video tutorials that will cover a lot of functionality. so stay tuned.

In the meantime, here's how the Ghost creation currently works. You can create a ghost sample that takes into account any arbitrary ancestry going back as many generations as you wish. Lets say you wanted to recreate a stand-in for someone who had some idea of the mixture(s) that led to their genesis. You can simply mix and match in the selection interface, and end up with a G25 sample that should produce similar results to this hypothetical individual of mixed ancestry. This sample can then be compared to the real individual for validation of their actual ancestry. For example, a Punjabi that had ancestry from 4 Punjabi communities might look like

https://nimbus-screenshots.s3.amazonaws.com/s/187f6dad2fb134298dca4b260ee8054a.jpg

In the next couple of days, we'll release an update that will allow for simple averaging across a set of selected samples

https://nimbusweb.me/box/attachment/4395932/edhzkcqnvcoht9t7qy38/lwrz5whg7skGFSEc/screenshot-localhost_8080-2020.07.03-21_11_51.jpg

Our goal is to continue to iterate rapidly as users like yourselves come forward with feature requests, so please keep them coming!

MonkeyDLuffy
07-04-2020, 06:36 AM
We're getting ready to release some video tutorials that will cover a lot of functionality. so stay tuned.

In the meantime, here's how the Ghost creation currently works. You can create a ghost sample that takes into account any arbitrary ancestry going back as many generations as you wish. Lets say you wanted to recreate a stand-in for someone who had some idea of the mixture(s) that led to their genesis. You can simply mix and match in the selection interface, and end up with a G25 sample that should produce similar results to this hypothetical individual of mixed ancestry. This sample can then be compared to the real individual for validation of their actual ancestry. For example, a Punjabi that had ancestry from 4 Punjabi communities might look like

https://nimbus-screenshots.s3.amazonaws.com/s/187f6dad2fb134298dca4b260ee8054a.jpg

In the next couple of days, we'll release an update that will allow for simple averaging across a set of selected samples

https://nimbusweb.me/box/attachment/4395932/edhzkcqnvcoht9t7qy38/lwrz5whg7skGFSEc/screenshot-localhost_8080-2020.07.03-21_11_51.jpg

Our goal is to continue to iterate rapidly as users like yourselves come forward with feature requests, so please keep them coming!

Thank you for detailed post. Also I wanted to bring an issue I ran into while using the Plot tool. Some populations do not show up even if you add them in the plot, I tried Afghan Pashtun and Simulated AASI, both didn't show up even though they show in the list. Is it a bug?

GenoPlot
07-04-2020, 05:26 PM
Thank you for detailed post. Also I wanted to bring an issue I ran into while using the Plot tool. Some populations do not show up even if you add them in the plot, I tried Afghan Pashtun and Simulated AASI, both didn't show up even though they show in the list. Is it a bug?

Not a bug, but there appears to be a mismatch between the local search cache and the PCA data. We'll get that corrected today.

vishankar
07-09-2020, 09:01 AM
Hi genoplot , i had asked this question in your help option ..received one reply and then no further correspondence...
so i am attaching the entire conversation here!.

""Your request (587) has been updated. To add additional comments, reply to this email.

Admin (genoplot.com)

Jul 2, 2020, 4:17 PM EDT

Hi,

Could you provide more detail on how you would like to organize your groups? Will these be public or private groups? And is the structure expected to be like:

Kerala > Group1
member1

member2

average (calculated by genoplot)

median (calculated by genoplot)

Kerala> Group2
member1

member2

member3

average (calculated by genoplot)

median (calculated by genoplot)


Or did you have a different structure in mind?

Thanks
.

Jul 2, 2020, 3:34 PM EDT

hi,
i have quiet a few coordinates cheifly belonging to communities from kerala state of india...how do i great groups and averages for these on Genoplot??
kindly enlighten.. regards!
ME.
------------------
Submitted from: https://genoplot.com/g25

Be a part of the Community: About Us | Help

You have been sent this message because you recently opened a help request with

[3QPPVR-P2QL]"

please help!....






my reply-
To:
Admin (Support)

Fri, 3 Jul at 1:26 pm

Hi...sorry for the late reply!.
yes..exactly like you have illustrated ...and including possible outliers!...i have more than 2 communities Data!...
Also....coordinates could keep getting added!...
Do you guys calculate coordinates from raw data?
when can i start sending coordinates ?
regards,

BalkanKiwi
07-09-2020, 10:58 AM
Hi genoplot , i had asked this question in your help option ..received one reply and then no further correspondence...
so i am attaching the entire conversation here!.

""Your request (587) has been updated. To add additional comments, reply to this email.

Admin (genoplot.com)

Jul 2, 2020, 4:17 PM EDT

Hi,

Could you provide more detail on how you would like to organize your groups? Will these be public or private groups? And is the structure expected to be like:

Kerala > Group1
member1

member2

average (calculated by genoplot)

median (calculated by genoplot)

Kerala> Group2
member1

member2

member3

average (calculated by genoplot)

median (calculated by genoplot)


Or did you have a different structure in mind?

Thanks
.

Jul 2, 2020, 3:34 PM EDT

hi,
i have quiet a few coordinates cheifly belonging to communities from kerala state of india...how do i great groups and averages for these on Genoplot??
kindly enlighten.. regards!
ME.
------------------
Submitted from: https://genoplot.com/g25

Be a part of the Community: About Us | Help

You have been sent this message because you recently opened a help request with

[3QPPVR-P2QL]"

please help!....






my reply-
To:
Admin (Support)

Fri, 3 Jul at 1:26 pm

Hi...sorry for the late reply!.
yes..exactly like you have illustrated ...and including possible outliers!...i have more than 2 communities Data!...
Also....coordinates could keep getting added!...
Do you guys calculate coordinates from raw data?
when can i start sending coordinates ?
regards,

I'm not involved with GenoPlot, but I can answer your last question. GP doesn't produce coordinates from raw data. If you're talking about G25 coordinates, you have to purchase them from the Eurogenes store in order to conduct modelling.

vishankar
07-09-2020, 05:17 PM
thanks @ balkankiwi...the Genoplot help/support system, has not been active for the past few days!...that is why i was forced to post here!

GenoPlot
07-09-2020, 05:34 PM
thanks @ balkankiwi...the Genoplot help/support system, has not been active for the past few days!...that is why i was forced to post here!

Apologies, we've been busy working on some new features for imminent release...

In the meantime, while we work out the mechanism for custom user defined sub-groups, we've gone ahead and added your custom Kerala groups for you along with their averages/medians:

"Kerala_Knanaya:mean"
"Kerala_Knanaya:median"
"Kerala_Nair:mean"
"Kerala_Nair:median"
"Kerala_Nambudiri:mean"
"Kerala_Nambudiri:median"
"Kerala_Payyanur:mean"
"Kerala_Payyanur:median"
"Kerala_Syrian_Christian:mean"
"Kerala_Syrian_Christian:median"
"Kerala_Thiyya:mean"
"Kerala_Thiyya:median"
"Kerala_Vaniya:mean"
"Kerala_Vaniya:median"

vishankar
07-09-2020, 06:47 PM
Apologies, we've been busy working on some new features for imminent release...

In the meantime, while we work out the mechanism for custom user defined sub-groups, we've gone ahead and added your custom Kerala groups for you along with their averages/medians:

"Kerala_Knanaya:mean"
"Kerala_Knanaya:median"
"Kerala_Nair:mean"
"Kerala_Nair:median"
"Kerala_Nambudiri:mean"
"Kerala_Nambudiri:median"
"Kerala_Payyanur:mean"
"Kerala_Payyanur:median"
"Kerala_Syrian_Christian:mean"
"Kerala_Syrian_Christian:median"
"Kerala_Thiyya:mean"
"Kerala_Thiyya:median"
"Kerala_Vaniya:mean"
"Kerala_Vaniya:median"

Dear Sir,
the Kerala _payyanur, needs to relabelled as payyanur poduval, hpe you will rectify that!...also i have plenty of fresh coordinates to add!...that will change the median and average right??

CyrylBojarski
12-08-2020, 01:41 PM
deleted