Hi Mohamed, 

I would  prefer to analyze the QTL mapping in each environment separately to see stability of QTL across environment. It will be a kind of testing the stability of QTL and validating in different environment.  Try to analyze QTL using both approaches (in combination and separately) and see the results how will they look like 

Hi Mohamed, 

I would  prefer to analyze the QTL mapping in each environment separately to see stability of QTL across environment. It will be a kind of testing the stability of QTL and validating in different environment.  Try to analyze QTL using both approaches (in combination and separately) and see the results how will they look like 

I agree with Ahmed, but would go a bit farther. If the ANOVA for your experiment indicates a significant GxE effect, it would be inappropriate to use the means across locations for QTL mapping. If you have data for more than 2 locations, you should conduct sub-analyses to explore which combinations of locations do not display significant GxE and can be grouped for QTL analysis.

Dear Mohammad

Genotype environment interaction can be detected by ANOVA of many genotypes in more than one environments , the significance of this source of variance in ANOVA , can confirm the presence of one of the two types of G×E interaction ( cross type or variance type ) .

Good Luck

Hi Mohamed,

Using combined data from multi environments is more powerful to map qtl with small effects and to detect the effect of each qtl in all environments which allows you to see whether the effect is from same allele or not. 

By using single environment analysis you loose the two advantages I mentioned earlier. 

I recommend multi environment analysis. 

For more information you can read the attached review.

Thank you dear all

But i want to keep open this discussion with asking this question, how heritability guides us to  implement  QTl analysis separately or by mean of data, while G*E is still significant.