Kinship Matrix

[Caesarius]

  Greetings Tassel Community,

How to interpret different MLM results  of Manhattan , QQ_plot ,when Centered vs Normalized kinship matrix is being used ?

There are 4 kinds of kinship matrixes that can be generated , but in a 6 year old language how would it be the best condensation key points of these.

  I am not sure what flavor of the kinship matrix is rTASSEL generating ?

  Working towards understanding ,

Caesarius

[Peter Bradbury]

The short answer is use the default, "Centered_IBS". That provides a reasonable estimate of an additive genetic relationship and is sometimes called a genomic relationship matrix (GRM). Roughly, a kinship matrix is calculated by multiplying the genotype score matrix by its transpose. In the case of Centered_IBS, the mean genotype score is subtracted from the values before multiplying. For the normalized version, the mean is subtracted then the result divided by the standard deviation. rTASSEL provides Centered_IBS.

Peter 

[Brandon Monier]

rTASSEL will default to Centered IBS, but users can provide different methods if they wish using the method parameter:

• Centered IBS ("Centered_IBS")
• Normalized IBS ("Normalized_IBS")
• Dominance Centered IBS ("Dominance_Centered_IBS")
• Dominance Normalized IBS ("Dominance_Normalized_IBS")

[Deeplearn Drone]

Thank you all for the high yield explanations !

Caesarius

[Deeplearn Drone]

Thank you, all  for the high yield insights !

Caesarius

On Friday, July 30, 2021 at 9:58:25 AM UTC-5 brandon...@gmail.com wrote: