Categorical histograms in seaborn from a dataframe

Question

Consider I have the following dataframe:

sample	tca_after	avg_length_after	cd_after	tca_before	avg_length_before	cd_before
1	0.015385	50.513499	1.4	0.005139	31.844415	0.4
2	0.005040	19.209373	1.0	0.004603	20.831459	0.6
3	0.057218	31.869649	10.0	0.008687	17.926937	1.0
4	0.037175	45.543659	3.8	0.035760	56.937708	2.8

I want to compare TCA, avg_length, and CD, before and after a certain process. So I would like to create three categorical histograms like that using seaborn. On x-axis I have all four samples, on y-axis I have either TCA, avg_length or CD for both before and after. I have no idea how to do it 🙁

Could you please help?

Asked By: Arman Bolatov

||

Source

Answer 1

You could join the before and after columns could together, and add a new column, e.g. "when" to indicate before vs after. That new column can then be used as "hue".

Here is how you could create a histogram:

import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd

df = pd.read_html('https://stackoverflow.com/questions/73348014')[0]

df_before = df[['sample', 'tca_before', 'avg_length_before', 'cd_before']]
df_before = df_before.rename(columns={'tca_before': 'tca', 'avg_length_before': 'avg_length', 'cd_before': 'cd'})
df_before['when'] = 'before'
df_after = df[['sample', 'tca_after', 'avg_length_after', 'cd_after']]
df_after = df_after.rename(columns={'tca_after': 'tca', 'avg_length_after': 'avg_length', 'cd_after': 'cd'})
df_after['when'] = 'after'
df_new = pd.concat([df_before, df_after]).reset_index()

sns.histplot(data=df_new, x='tca', hue='when', palette='copper_r', multiple='dodge')

To create a bar plot with the sample IDs on the x axis, you can change sns.histplot to sns.barplot:

sns.barplot(data=df_new, x='sample', y='tca', hue='when', palette='copper_r', dodge=True)

Answered By: JohanC

Categorical histograms in seaborn from a dataframe

Question:

Answers: