For phenotyping disease resistance/susceptibility in a plant, we have digitally scanned the leaf area of many leaves over a time period after inoculation (controlled environment). Would using a disease scale by which to categorize disease severity and calculate a DSI be more accurate than simply taking the average disease severity of samples at the given time period?
s. Jacob Schneider 1) Scoring the DSI depends on personal experience. Even with extensive training with computer - based tests, two researchers will put different scores for the same set of plants. Once we made such experiment on 120 experimental plots with Septoria on wheat (50 plants per plot, 4 leaves per plant, 4 times for season). The difference between DSI from two estimators was significantly above standard errors for each estimator.
2) Distribution of DSI can be different from the scanned diseased leaf area. DSI is more ranking than parametric distribution (I am not sure that it is statistically correct).
I am not sure I understood the question correctly, but using any kind of scale which is based on infected surface area instead of the infected surface area itself, only means you lose data by this, which definetly is a big no (unless you want to obtain the data faster). It is even a common strategy to calculate back the scale data to percentages, based on the midpoints of the scale category. If you have multiple assessment date during a period instead of average try AUDPC.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication