PDA

View Full Version : Mathematical Formulae used in genetic genealogy



Michael Clarke
12-18-2015, 10:54 PM
There seems to be an increasing use of various mathematical formulae in genetics, which are difficult for the non-mathematician to understand - FS3, D stats, etc. Can anyone point to an online resource which explains to the layman how they work, and how to interpret the outputs?

anglesqueville
12-18-2015, 11:30 PM
There seems to be an increasing use of various mathematical formulae in genetics
Without mathematics (essentially rather elementary probabilities theory) and statistics, there would be no population genetics at all. The increasing use of mathematics in population genetics is only the signal that this discipline is become fully scientific, far from the phantasms of the XIXth century.
About f2- f3- D-stats, one of the founding texts : http://www.genetics.org/content/early/2012/09/06/genetics.112.145037.full.pdf+html . Probably unreadable if you have no mathematical background, but you can try to understand the examples. If you want more, I don't see other alternative but to buy a good undergraduate book of probabilities and statistics, and to work hard.

Michael Clarke
12-19-2015, 06:08 PM
Thanks for the reference which I will work through.

SveinD
12-19-2015, 07:13 PM
Without mathematics (essentially rather elementary probabilities theory) and statistics, there would be no population genetics at all. The increasing use of mathematics in population genetics is only the signal that this discipline is become fully scientific, far from the phantasms of the XIXth century.
About f2- f3- D-stats, one of the founding texts : http://www.genetics.org/content/early/2012/09/06/genetics.112.145037.full.pdf+html . Probably unreadable if you have no mathematical background, but you can try to understand the examples. If you want more, I don't see other alternative but to buy a good undergraduate book of probabilities and statistics, and to work hard.

I found this article far to theoretical, but that could be me! A far more practical approach is found here: http://home.earthlink.net/~odoniv/HamCountry/HAM_DNA_Project/HAM_DNA_Phylogenetic_chart_Instructions.html

Using the 3 programs I have managed to generate many phylogenetic trees based on STR test data. Be warned that if you want to process a large database, i.e say greater than 75 results you need a fast processor - or be very patient!