Clarification on Tassel Tutorial on MLM

[K K Vinod]

Hi

Following is the text (in red) copied from the Tassel Online tutorial at https://bitbucket.org/tasseladmin/tassel-5-source/wiki/UserManual/AssociationanalysisusingMLM/AssociationanalysisusingMLM. I have tried to rerun this analysis using the Tassel Version 5.2.19 using the TutorialData. However I am not getting the table as given blow in the Tutorial. No matter, I am unable to follow what is written below. 

a. Does this mean that even the strongest SNP based on the P value can go non-significant?

b. How do you identify the SNP  non-significant when the P is below threshold?

c. The Illustration shows three allellic states of the highlighted marker, is this the reason for claiming the association as non-significant?

Please anyone help me. 

I find lot of variation between the Tutorial and the Tassel Version 5. When will we have an updated version of the Tutorial?.

Thanks in advance

V

The strongest associated SNP is at 193565357 bp on chromosome 3. The P value is 1.3027x10-4. The threshold is 3.2331x10-5 at significant level of 1% after Bonferroni multiple test correction (0.01/3093). The association was not significant. As illustrated below, the output labeled “MLM_effects_for...” shows the marker effects assigned to genotypes for each SNP (The GLM is also the same). For example, the first SNP at 157104 bp on chromosome 1 had three genotypes (AA, CC and AC) coded as A, C, and M based on the IUPAC code (see Appendix).

[K K Vinod]

Any help? Thank you!

[Peter Bradbury]

a. yes, the strongest SNP is not significant. 

b.3000 SNPs is not enough for association analysis in a diverse maize panel. Our expectation was that no SNPs would be significant. Only p values smaller than the Bonferroni threshold should be considered significant. The p-value in question is larger.

c. No. The null hypothesis is that all three genotype classes are equal. The Bonferonni corrected test fails to reject that hypothesis. 

more detailed information is available int the user manual under the analysis section (https://bitbucket.org/tasseladmin/tassel-5-source/wiki/UserManual/MLM/MLM)

We have not been putting much effort into updating the tutorial section of the User Manual recently, as you noticed. However, a member of our lab has been making tutorial videos, which are available through You Tube at https://www.youtube.com/channel/UCS1SdXlyMI1OsSf5yA_oFqw or from a link on the Tassel page at maize genetics.net.

Peter

[K K Vinod]

Thanks, Peter!

Vinod

[carolina.a...@gmail.com]

Hi,
Could you explain what is the exact definition of the estimate of the effect of each allele?
Thank you

On Wednesday, February 3, 2016 at 8:38:05 PM UTC-8, K K Vinod wrote:

Thanks, Peter!

Vinod