Post imputation checklist

[megan-soria]

• [x] make a new df from original GO input matrix with blinded genes = 0 GOs

• [x] make the same correlation analysis

maybe do this for 10 clusters

• [x] Do the comparison for cluster 1 and check accuracy and precision by matrix subtraction (blinded df- original df)

• [x] tweak parameters: (1) correlation threshold, (2) number of clusters (*the higher accuracy, the better), and/or (3) use another clustering method like K-Means

[megan-soria]

task 4 related to commit#: a952aa59b407d4895e013926eaa6595648bee326