I would suggest using R, it is free and lot of people use it. However, if you are willing to pay for a something SAS or Matlab would be best. 

I would suggest looking at R and bioconductor packages to see if they have what you need. (Possible packages that might work for you would be diveRsity or QTLRel it depends on your input data.)   

I am not sure if this is what you are looking for but there is a step by step process for a simple example for calculating genetic coefficients in R.

https://ecoqui.wordpress.com/2015/03/06/how-to-calculate-genetic-variance-components-coefficient-of-genetic-variation-and-genetic-correlations-in-r/ 

There are no absolutely correct answers to your question as it depends on what you are trying to do. 

 I would agree with the previous post. R is probably the best statistical suite for such work in general. The downside of most R programmes is that the documentation can take some work.  

SAS and similar programmes are very useful but can be expensive if you only need them occasionally. 

If you data set is very large R may not be the best choice as it can be quite slow. If the data set is very large and you know the formula you wish to apply, a home brew in C is probably the fastest way to obtain the estimate you are looking for. 

 If you need estimates of various parameters in your formula OpneBugs might be worth a look. This is a Bayesia system. While powerful it can be slow. Using it to estimate parameters of a formula may be overkill but it is a useful way of getting used to a powerful (if slow) system. 

Many thank to both of you. I have solution of my concern. Special thank to Anna for the link

Dear Zaki

I think the Meta-R will be helpful for you: http://data.cimmyt.org/dvn/dv/cimmytswdvn/faces/study/StudyPage.xhtml;jsessionid=a639be3a3102134e1fb4d393e043?globalId=hdl:11529/10201&studyListingIndex=0_a639be3a3102134e1fb4d393e043