Minor issue #1: When you're reading in the eigenvectors for the PCA plot, you never convert the values to floats (they are strings by default), so you're plotting strings, which is why your PCA plot looks so weird. For some reason, matplotlib doesn't through an error when you try to plot strings, it just does its best, which is wrong.
Minor issue #2: In your manhattan plot, you're plotting all of the significant points on the left of the plot, rather than in the appropriate spots relative to the other variants. When you extract the significant variants, you also want to extract their x-values, which can just be their index in the full list of variants.
Pretty plots
4/4
Exercise
Points Possible
Grade
Step 1.2 PC plot
1
1
Step 2.2 AFS plot
1
1
Step 3.2 Manhattan plots
1
1
Step 3.3 effect size boxplot
1
1
Grade
Total: 9.25/10
Great work! Just fix the minor issues and resubmit for the full 10/10
README.md
with commands and analyses2/2
plotting.py
script to produce plots3.25/4
Minor issue #1: When you're reading in the eigenvectors for the PCA plot, you never convert the values to floats (they are strings by default), so you're plotting strings, which is why your PCA plot looks so weird. For some reason, matplotlib doesn't through an error when you try to plot strings, it just does its best, which is wrong.
Minor issue #2: In your manhattan plot, you're plotting all of the significant points on the left of the plot, rather than in the appropriate spots relative to the other variants. When you extract the significant variants, you also want to extract their x-values, which can just be their index in the full list of variants.
Pretty plots
4/4
Grade
Total: 9.25/10
Great work! Just fix the minor issues and resubmit for the full 10/10