PCA Analysis to cluster experiments using H3K27ac signal mouse enhancers, taking log of signals + small pseudo coun and then create a scatter plot.

0

Entering edit mode

2.9 years ago

Simaranrajorass22 • 0

PCA Analysis to cluster experiments using H3K27ac signal mouse enhancers, taking log of signals + small pseudo count and then create a scatter plot.

I ran PCA analysis but getting complex numbers in eigen vectors and thus I am not able to plot a graph. Any idea on how to avoid complex numbers or dealing with such data.

I normalized my data using: I took log2(df+1.025)

How exactly can I normalize and shorten my data to avoid complex numbers?

t-SNE PCA UMAP • 847 views

ADD COMMENT • link updated 2.9 years ago by Mensur Dlakic ★ 28k • written 2.9 years ago by Simaranrajorass22 • 0

0

Entering edit mode

What do you mean by complex numbers exactly? Are you getting an error?

ADD REPLY • link 2.9 years ago by jared.andrews07 ★ 18k

0

Entering edit mode

How about standardizing the data by subtracting the mean? Or by subtracting the mean and scaling by standard deviation?

ADD REPLY • link 2.9 years ago by Mensur Dlakic ★ 28k

Login before adding your answer.