Chapter 9 Final Takeaways

In general, I feel like the sW22 lineage influences most of the traits measured. However, it seems to be not as clean as I would have expected. There isn't always a reduction in the early generations that continues to decline, sometimes there's no difference between the mean of the control (W22, B73, or Mo17) at one generation and then it appears at the next.

For many of these traits, it does seem like the trait is moving towards the initial value. I'm not sure if I'm sold on the idea that there is a continues decline in these traits as we keep backcrossing. I also suspect that there is an impact for the sex of the sick parent. In general, I found that the paternal sick parent had higher values than the maternal sick parent, but it's variable.

I'm also a little concerned about the low sample sizes for some of these traits, especially when we separate them by field. For the 2019 data, many of the comparisons in field 2 only had 5 values. I think scaling things up, taking into account that we're likely going to have missing data at the end due to the plants being sickly, may be a good idea.

Overall, I'm honestly just more confused about the exact nature of what's going on here. Clearly something is affecting the sW22 crosses in a way that the W22 crosses don't seem to be, but it's just not a consistent picture to me yet.

9.1 sW22 backcrossed to W22

For most of the phenotype data collected, we find that there is a decrease in the trait of interest in the F1 or early generation backcrosses. This recovers by the later generation backcrosses for the plant and ear height phenotypes, but for the others, ear length, seed weight, cobb weight, and estimated seed number, there seems to be a constant reduction. We also find that if there is a consistent difference in the phenotype when comparing the sex of the sick parent, the paternal crosses (where thte paternal parent is sick) tend to have higher/heavier values than the maternal crosses (2019 Plant Height, 2019 Ear Height, 2019 Seed Weight, 2019 Cobb Weight, and 2019 Estimated Seed Number). When we specifically test for the significance of the sex of the sick parent on these phenotypes for the sW22 cross, we find that there is a significant (often highly significant) influence of the sex of the sick parent.

9.2 sW22 and W22 backcrossed to B73

For the sW22_B73 backcrosses, we usually see either an increase or no change in the phenotype trait that slowly declines as we continue to backcross to B73. For plant height, this trend holds out very nicely in the 2020 height data but not as well in the 2019 data. It is also nicely demonstrated by ear length, seed weight, cobb weight, and estimated seeed number. For most of these crosses, there is not a consistent difference between crosses depending on the sex of the sick parent. The only ANOVA where this is found to be significant is for Estimated Seed Number. However, for all of the phenotypes except for ear height, the ANOVA finds that phenotype is signficantly influenced by the segregating genotype.

When we compare the trait data at each generation to the matched W22 to B73 cross, we almost always find that the trait is reduced compared to the W22_B73 cross.

9.3 sW22 and W22 backcrossed to Mo17

Similar to the sW22_B73 crosses, we find that the sW22_Mo17 crosses usually show an increase in the trait followed by a decline. However, there are a few phenotypes that seem to be insignificant (i.e. there doesn't seem to be a significant difference between the trait in the backcrosses and MO17) including 2019 plant height, 2019 ear height, and 2019 ear length. For plant height, ear height (if the cross is sick paternal), and cobb weight we do not find that the sW22_Mo17 value is significantly different than the W22_Mo17 value.