Noise Threshold On Autosomal DNA IBD Segment Matches

[J Man]

This is quite an interesting explanation/take in the noise threshold for autosomal IBD segment matches. What do others here think about this guy's interpretation?

http://www.y-str.org/noise-threshold-on-atdna-matches/

[AJL]

This guy doesn't know what he is talking about.


After doing some research, I found that any segment length even on an unphased DNA which is greater than 3 cM is an IBD

What research? Research of others shows that fake IBD segments have been routinely observed at 5 and 6 cM, and sometimes as large as 10 cM.


The entire women of a town are often raped by the invading army in those days after killing every male.

Really?


If the population is 7 billion, with 10 genotypes, there are 700 million people for each genotype.

False. You can only derive SNP rates empirically, not by making up figures Mathematically. Rates vary a lot from SNP to SNP and population to population.


Proof using Cryptography

You cannot compare a manmade code where each cypher character is equally probable to a genetic code where different letters code for different functions. In a cypher, "0" and "1" are neutral in their effects: in an amino acid, "A" and "C" can cause very different effects in an organism.

[J Man]

This guy doesn't know what he is talking about.



What research? Research of others shows that fake IBD segments have been routinely observed at 5 and 6 cM, and sometimes as large as 10 cM.



Really?



False. You can only derive SNP rates empirically, not by making up figures Mathematically. Rates vary a lot from SNP to SNP and population to population.



You cannot compare a manmade code where each cypher character is equally probable to a genetic code where different letters code for different functions. In a cypher, "0" and "1" are neutral in their effects: in an amino acid, "A" and "C" can cause very different effects in an organism.

Interesting thanks for your thoughts AJL.