Checklist For Setting Up a Phenotyping / MAB Study
There are some key points which you should keep in mind when working with MAB and phenotyping research.
- Get a statistician to advise you - before you start your experiment. After you do your phenotyping and try to analyze your data, it is too late to go back and design a good field experiment!
2. Don’t guess! It is better to have missing data than wrong data (Illustration 5)
A missing data point is not the same as a score of 0. For example, with yield: a score of 0 grams is much worse than an unknown yield. A score of 0 means that the plant did not yield anything (the genotypic make-up of that plant results in no yield) while a missing plant (perhaps did not germinate or died while in the field) may have genes associated with high yield, but we can’t tell.
The final format your data will need to be in depends on which software program you will use to analyze it, but it is often in a format similar to Illustration 7, where the plant names are given in the first column, and the trait names are given in the first row. Here, a “-” represents missing data. This is in Excel.
Figure 6/Illustration 5: Data files-correct entry of data is key. Here is an example of 30 tomato plants with various trait measurements (Bost = Bostwick, a viscosity measurement; Brix = a sugar or solids measurement; FtWt. = fruit weight; ab = fruit color measured by spectrophotometer). Notice that missing data is scored as “-“ so as to differentiate it from 0. (Data from Theresa Fulton)
3. Data organization is key. An important thing to note is that trait data need to be arranged in exactly the same order as the marker data were. Your field plots will be randomized, so you need to make sure you have re-arranged them to fit the marker data before you start to analyse the whole data set.