Resources

Talks

Heritability:
Summary of heritability equations for balanced and unbalanced datasets.
There are more ways to calculated H2 based on if you do spatial correction, gxe, and so on, but this is a start.

You can obtain σ g 2 (genotype variance) and σ e 2 (residual variance) in R using the lme4 package with VarCorr(model) after you’ve defined your model, but Dan Sweeney shared with me his function to calculate vBLUP:

  
library(lme4)
model1 <- lmer(Pheno~(1|genotypes)+Env+HD, data = EliteProgramData) 
S_1 <- as.data.frame(VarCorr(model1)); 
sigma2_g <- S_1[1,4]
sigma2_e <- S_1[2,4] #Assuming your model only has 1 random effect. If more than one, [2,4] becomes [n,4] where n = number of random effects + 1 

#Balanced H2
FalMac_H2 <- sigma2_g / (sigma2_g + sigma2_e)

m <- length(unique(EliteProgramData$Env))
MultiEnv_H2 <- sigma2_g / (sigma2_g + (sigma2_e/m))

#Unbalanced H2
Cullis_H2=function(model){
  library(arm)
  ses<- se.ranef(model)$'genotypes' #where 'm' is your model object from 'lmer' (replace 'genotypes' with whatever you call your individuals in the data)
  v_BLUP<- ses^2
  sigma2_g=VarCorr(model, comp="Variance")$'genotypes'[1]
  Reliability<- 1- v_BLUP/ (2\*sigma2_g)  #where sigma2_g is the genetic variance estimated with the model saved in 'm'
  H2<- round(mean(Reliability),3) #This is equivalent to broad-sense heritability on the line-mean (or family-mean, if your individuals are non-inbred families) basis
  H2
}

Cullis_H2(model1)

90k Illumina Array SNP Positions

Quite often, as I read a mapping journal article, I want to know if a particular Illumina 90k marker linked to the published QTL. Originally I had to blast the marker one at a time against the wheat genome, but that gets tedious (and you also need to know the marker sequence, fun right). But URGI does have the SNPs aligned to the RefSeqv1.0 JBrowse. So, I downloaded the Infinium 90k track data for every chromosome and compiled them into one 90k position excel file for a quick reference. Feel free to download the quick resource for easy comparison of marker with position.

BUT, another issue I often run into is only knowing the IWB#### index number OR the marker name (ex: BobWhite_c10090_559) when I need the other number/name. So, here is another 90k information combined file that includes alignment of the IWB number, the marker name, flanking SNP sequences, the A allele nucleotide, the B allele nucleotide, and even the original 9k_name if appropriate.

The Falling Numbers Website

PHS/Dormancy Genes Positions
A number of genes in wheat have been cloned and characterized for PHS tolerance or grain dormancy. Since the wheat reference genome is now publically available, we can BLAST the PHS gene sequences (below) against the RefSeqv1.0 alignment and identify the positions on the chromosome. Below are the estimated start and end positions based on the gene sequences available by the authors, in the publication, or public databases. The positions and links to the gene sequences may be useful to cross-reference with future PHS QTL or GWAS analyses.

Gene		Chrm	Pos_Start	Pos_End	RefSeqv1.0_Gene_Name	Reference	Gene_Seq
TaSdr / Sdr	TaSdr-A1	2A	158,452,164	158,453,676	TraesCS2A01G191400	Zhang et al., 2014	KF021988
	TaSdr-B1	2B	200,574,062	200,572,560	TraesCS2B02G215300		KF021990
	TaSdr-D1	2D	142,668,463	142,666,958	TraesCS2D01G196200		KF021992
Tamyb10 / R	R-A1	3A	703,906,112	703,904,221	-	Himi et al., 2011	AB599721
	R-B1	3B	757,917,663	757,920,399	TraesCS3B01G515900		AB191459
	R-D1	3D	570,799,694	570,803,366	TraesCS3D01G468400		AB191460
TaVp-1 / Vp1	TaVP-1A	3A	659,556,434	659,552,099	TraesCS3A01G417300	Yang et al., 2014	Vp-1A
	TaVP-1B	3B	693,342,691	693,338,300	TraesCS3B01G452200		AJ400713
	TaVP-1D	3D	525,473,897	525,469,502	TraesCS3D01G412800		AJ400714
TaPHS1 / TaMFT	TaMFT-3A	3A	7,294,362	7,297,912	TraesCS3A02G006600	Nakamura et al., 2011	G1UE17
TaDOG1	TaDOG1L4	3A	67,123,800	67,124,332	TraesCS3A01G103500	Ashikawa et al., 2013	AB555729
		3B	91,147,446	91,147,986	TraesCS3B01G120900
		3D	58,107,736	58,108,268	TraesCS3D01G105800
TaMKK3 / Phs-1	TaMKK3-A	4A	605,019,000 a	605,023,000	TraesCSU01G167000	Torada et al., 2016	LC091368
	TaMKK3-B	5B	710,226,100	710,230,860	TraesCS5B01G565100
	TaMKK3-D	5D	556,557,389	556,553,063	TraesCS5D01G549600		LC091372
TaQsd1 / Qsd1 b	TaA	5A	432,446,370	432,446,370	TraesCS5A01G216200	Onishi et al., 2017	Qsd1
	TaB	5B	387,744,682	387,740,080	TraesCS5B01G214700
	TaD	5D	332,043,225	332,043,225	TraesCS5D01G224200

a RefSeqv1.0 maps TaMKK3-A to the unknown chrm, however based on previous research, TaMKK3-A should be between TraesCS4A01G314700 and TraesCS4A01G314800
b the ortholog of the barley dormancy gene HvQsd1. The Onishi et al. 2017 study compared sequence of dormant and non-dormant lines, however further work needs to be published to see if TaQsd1 genes affect seed dormancy