I suggest you handle your data with winQTLCart 2.5. Firstly, you need to combine the phenotype data with genetic data; then, you can choose QTL mapping methods according to your datasets. Single marker analysis is quite simple. Good luck!

If you are just doing single marker analysis, it is essentially a regression where the phenotype is dependent upon the value of the genotype (generally coded 0,1 for dominant markers or -1,0,1 for codominant markers). You can easily do the calculations in Excel (or with paper and pen, if you have time to kill...).

There's a good example at the link I provide below.

If you have a more complicated experimental design, you can expand the regression model into an ANOVA by adding additional terms (blocks, years, locations, etc.)

If you are going to build a linkage map using the marker info, then I would suggest doing as Hui-jie Zhai recommends and use QTLCartographer or R/qtl to perform interval or composite interval mapping, as appropriate.

https://articles.extension.org/pages/32523/application-of-anova-for-plant-breeding:-single-marker-analysis-example-using-anova-in-a-balanced-po

For single marker analysis, you could use ANOVA as suggested by Jack or Kruskal-Wallis (KW)Test (without worrying about the data normality). You could consider Joinmap for building your linkage map and MapQTL for subsequent QTL mappings. KW test is included in the second program. However, you need to pay for these programs once for life-time license.

Do you have specific reason to use single marker analysis? As suggested by Jack, interval and composite interval mapping are more powerful.   

Dear ,  if you are going to conduct validation study of detected QTLs or like to detect QTL in population like BCnFn, you can use SMA, & you conduct ANOVA test as mentioned by  Jack. In case of higher No. of samples, you can use QTL cartographer

And, in case you are going to conduct QTL mapping (Linkage) in F2, F3, or RILs, it is better to use IM, CIM, or MIM. Thanks