Yelling - is it child abuse?

[pahoben]

That's kinda scary really.

You can request anonymity in most services but many people choose not to. It is kind of an ingenious use of public data. Another atypical use is adopted children trying to identify their biological parents. I understand there are sometimes problematic results like when siblings provide samples and find they have different fathers even though their mother has been married to one man. Counseling ensues.

 

[amberpi]

Totally agree with others though-y'all gotta be on the same page on things

DH makes the rules, but we've never disagreed, or at least not yet. There are things we talk about but so far we have the same values. Wish me luck it continues.

 

[pahoben]

Are you suggesting the ancestry databases are analyzing the full genome? If so, that does not fit with expert testimony I've heard from forensic biologists who perform DNA analysis and some casual conversations I've been involved in with some of them. The algorithms necessary to come up with results in probabilistic genotyping are incredibly complex simply to determine the likelihood that the comparison DNA from a known sample is either not likely to be a match for an unknown donor on a specific piece of evidence or X sextillion times more likely to be a donor to that sample than an unrelated, unknown donor (that's how the latest DNA analysis method expresses those types of results, the awkward wording is not my lack of skill at expressing the concept.)

Comparing a known reference sample they take from me as a potential donor and comparing it to an unknown DNA profile extracted from a piece of evidence only looks at a few locations on my known DNA profile and the same few locations on the unknown DNA profile extracted from the piece of evidence. The entire genome is not analyzed to do this.

Simply the economics of the ancestry databases analyzing every location on the entire genome makes it incredibly unlikely. Attempting to extract the probabilistic genotyping for the specific behavior traits as you've suggested from an analysis of the entire genome doesn't seem feasible from my understanding of how DNA analysis is conducted. As I understand it the best way to obtain results such as you propose would be to focus analysis only on the data from specific locations on the genome. My thoughts also correlate with the genetic data you expressed upthread identified with letter/number identifiers. I believe those are meant to indicate specific locations on the DNA profile where the data was taken from.

The ancestry sites wouldn't have a need to focus on those specific locations and therefore wouldn't be utilizing resources to analyze that. If they're not conducting the analysis, who is? (And who's paying for it?) I'm not sure of the particulars of how far the legal release of information stretches when giving a sample to an ancestry DNA site, but in any case it seems to me it takes more than access to the profile to connect the dots when claiming a genetic component to a behavioral trait. Without the ability to take the results of analysis of relevant locations on the profile and then confirm them with the human donor, how can one say definitively that this specific gene indicates this person does in fact exhibit the behavioral trait in question? In order for that comparison to take place it means that personally identifying information must be included in the release of information.

Here is a service that offers complete 100% sequencing for $999 but just had a sale for $200. I have a large part of my genome sequenced including full sequencing of my Y chromosome as part of an FTDNA research project. There is a lot of behavioral genetic research being conducted at this time by both public and private universities and some funded by NIH. So I guess in part taxpayers are paying for much of this.

These data bases are not for legal use ostensibly but just general interest to participants and use for ancestral purposes.

If certain polymorphisms say have a frequency of .25 in the general population and two samples match at 20 polymorphism then that would happen by chance say one in a trillion times so if the court accepts that as sufficient evidence that they are from the same person then case closed i guess but that is not the purpose at all for participants in these data bases. It is more curiosity and also to allow genetic research to proceed and for genealogical purposes.

 

[pahoben]

Sorry. Here is the link. I believe they are associated with Carnegie Mellon

https://www.veritasgenetics.com/myGenome

 

[pahoben]

Sorry. Here is the link. I believe they are associated with Carnegie Mellon

https://www.veritasgenetics.com/myGenome

The human genome is 3 billion bases long and 2 bits/base and 8 bits/byte so call it 750 megabytes-much smaller than a Blu-ray movie for storage requirements.

 

[cabanafrau]

Sorry. Here is the link. I believe they are associated with Carnegie Mellon

https://www.veritasgenetics.com/myGenome

This is completely different from the ancestry databases we were discussing -- one of the very first things this website talks about is how they are different from 23 and Me and the like. It specifically references how something like 23 and Me looks at about 1% of your DNA, as I suggested. 23 and Me and their competitors are not looking for the kind of behavioral traits you suggested were being mined from the growing databases of the ancestry sites. What you're linking to here costs a lot more because it encompasses much more beyond the ancestry DNA sites. I'll stand by what I've said previously.

 

[pahoben]

The human genome is 3 billion bases long and 2 bits/base and 8 bits/byte so call it 750 megabytes-much smaller than a Blu-ray movie for storage requirements.

I guess some overhead for labeling the chromosomes that each base lies on but even at one byte/base still at 3 gigabytes and still less than a blu ray movie.

 

[pahoben]

No they are not-other research groups are looking at the links. I don’t understand. Are you claiming that this work is not being done. Do you want me to provide links to specific research papers?

 

[cabanafrau]

I guess some overhead for labeling the chromosomes that each base lies on but even at one byte/base still at 3 gigabytes and still less than a blu ray movie.

Yeah, nothing to it. Simple algorithms.

 

[pahoben]

You can easily find many papers with google search.

 

[soccerdad72]

I guess some overhead for labeling the chromosomes that each base lies on but even at one byte/base still at 3 gigabytes and still less than a blu ray movie.

How many more times are you going to quote yourself?

 

[pahoben]

By the way-that is not an algorithm but only a non interactive calculation.

 

[cabanafrau]

Research has been conducted since the 70’s and with twins at that time both fraternal and identical that providing an avenue to study genetics and behavior (sorry about my reference to my twins and I just wanted to supply observations that many people do not have since they do not have twins-my mistake) and those studies found links between genes and behavior. Now we have available large genetic genetic data bases ( FTDNA 23 andMe etc). to test this in a different way. These analyses
Have supported the twin studies and have even located specific genes that correlate with personality traits. It isn’t that I thought you couldn’t understand but this is a topic that some people have deep beliefs about and will never ever change their beliefs.

No they are not-other research groups are looking at the links. I don’t understand. Are you claiming that this work is not being done. Do you want me to provide links to specific research papers?

You raised the issue suggesting the studies were being done with data gleaned from the growing databases of ancestry sites -- specifically mentioning 23 and Me, etc. I said that those ancestry sites weren't studying those locations on the DNA profiles.

Now you've pivoted to a completely different site doing an analysis that a pathetically lackluster single click immediately delineates as being much more comprehensive than 23 and Me, etc.

 

[pahoben]

By the way-that is not an algorithm but only a non interactive calculation.

One more time

Iterative not interactive.

 

[cabanafrau]

By the way-that is not an algorithm but only a non interactive calculation.

I shall alert the forensic biologists who earn their keep performing DNA analysis that they don't know what they're talking about.

 

[pahoben]

You raised the issue suggesting the studies were being done with data gleaned from the growing databases of ancestry sites -- specifically mentioning 23 and Me, etc. I said that those ancestry sites weren't studying those locations on the DNA profiles.

Now you've pivoted to a completely different site doing an analysis that a pathetically lackluster single click immediately delineates as being much more comprehensive than 23 and Me, etc.[/
I guess you don’t want links to any of the papers.
Here is a link to use by a police agency using frdna that successfully identified a suspect through gedmatch
https://www.google.ru/amp/s/www.theatlantic.com/amp/article/561695/

Also there are only about 20,000 genes in the human genome that comprise about 1% of the genome. The remainder is non coding DNA. If you are looking at the correlation between genes and behavior you only need to look at 1% of the genome.

I am really struggling to see the point of this so thanks for your kind words.

 

[cabanafrau]

Another pivot.

I never said people cannot be identified via ancestry site databases. That is very different data than that which seeks to genetically identify behavioral traits. It would make absolutely no sense to say that ancestry DNA databases couldn't identify people.

 

[pahoben]

You raised the issue suggesting the studies were being done with data gleaned from the growing databases of ancestry sites -- specifically mentioning 23 and Me, etc. I said that those ancestry sites weren't studying those locations on the DNA profiles.

Now you've pivoted to a completely different site doing an analysis that a pathetically lackluster single click immediately delineates as being much more comprehensive than 23 and Me, etc.

I guess you don’t want links to any of the papers.
Here is a link to use by a police agency using a commercial data base that successfully identified a suspect through gedmatch
https://www.google.ru/amp/s/www.theatlantic.com/amp/article/561695/

Also there are only about 20,000 genes in the human genome that comprise about 1% of the genome. The remainder is non coding DNA. If you are looking at the correlation between genes and behavior you only need to look at 1% of the genome.

I am really struggling to see the point of this so thanks for your kind words.

 

[pahoben]

Another pivot.

I never said people cannot be identified via ancestry site databases. That is very different data than that which seeks to genetically identify behavioral traits. It would make absolutely no sense to say that ancestry DNA databases couldn't identify people.

Human genes comprise only 1% of the genome and hence to look at links between genes and behavior you only need that 1% of the genome so in fact you were wrong in stating 23and me had insufficient data to study these links since only 1% there. In fact 1% is all you need.

 

[cabanafrau]

I guess you don’t want links to any of the papers.
Here is a link to use by a police agency using a commercial data base that successfully identified a suspect through gedmatch
https://www.google.ru/amp/s/www.theatlantic.com/amp/article/561695/

Also there are only about 20,000 genes in the human genome that comprise about 1% of the genome. The remainder is non coding DNA. If you are looking at the correlation between genes and behavior you only need to look at 1% of the genome.

I am really struggling to see the point of this so thanks for your kind words.

I've already answered this above in post 137 as I could not quote what you wrote in 136.

No, I don't want the link to any papers. I don't think I can explain what I have said any clearer for you and I have no desire to keep repeating my answers when you move the goalposts yet again.