Four treatments (Capomulin, Infubinol, Ketapril, and Placebo) for comparison. tasked with:
- Creating a scatter plot that shows how the tumor volume changes over time for each treatment.
- Creating a scatter plot that shows how the number of metastatic site changes over time for each treatment.
- Creating a scatter plot that shows the number of mice still alive through the course of treatment
- Creating a bar graph that compares the total % tumor volume change for each drug across the full 45 days.
#Observations:
- Ketapril actually does more harm than good. The mice who received it did worse than the mice who received nothing.
- Capomulin is the only one of the four who reduced the tumor size, number of metastatic sites and had the best survival rate.
- Looking at all of the drugs, Ramicane was as successful as at Capomulin, and they are the only two in the entire study with positive results.
# Dependencies
import os
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from scipy import stats# Read data
trial_data = pd.read_csv("raw_data/clinicaltrial_data.csv")
mouse_data = pd.read_csv("raw_data/mouse_drug_data.csv")#load the data into dataframes
trial_df=pd.DataFrame(trial_data)
mouse_df=pd.DataFrame(mouse_data)#check for duplicate mouse data and clean it prior to merging the datasets
duplicates = mouse_df.set_index('Mouse ID').index.get_duplicates()
bad_mouse = duplicates[0]
clean_mice=mouse_df[mouse_df['Mouse ID'] != bad_mouse]
clean_trial=trial_df[trial_df['Mouse ID'] != bad_mouse]
dataset=pd.merge(clean_mice,clean_trial, on = "Mouse ID")
dataset.head()
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Mouse ID | Drug | Timepoint | Tumor Volume (mm3) | Metastatic Sites | |
|---|---|---|---|---|---|
| 0 | f234 | Stelasyn | 0 | 45.000000 | 0 |
| 1 | f234 | Stelasyn | 5 | 47.313491 | 0 |
| 2 | f234 | Stelasyn | 10 | 47.904324 | 0 |
| 3 | f234 | Stelasyn | 15 | 48.735197 | 1 |
| 4 | f234 | Stelasyn | 20 | 51.112713 | 2 |
#table of tumer size over time per drug
tumor=dataset[['Mouse ID', 'Drug', 'Timepoint', 'Tumor Volume (mm3)']]
tumor_data=pd.pivot_table(tumor, index= ["Drug", "Timepoint"] )
tumor_data.head()
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Tumor Volume (mm3) | ||
|---|---|---|
| Drug | Timepoint | |
| Capomulin | 0 | 45.000000 |
| 5 | 44.266086 | |
| 10 | 43.084291 | |
| 15 | 42.064317 | |
| 20 | 40.716325 |
#table of all drugs over timepoints
tumor_table=tumor.pivot_table( index= "Timepoint", columns = "Drug", values = "Tumor Volume (mm3)")
tumor_table
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Drug | Capomulin | Ceftamin | Infubinol | Ketapril | Naftisol | Placebo | Propriva | Ramicane | Stelasyn | Zoniferol |
|---|---|---|---|---|---|---|---|---|---|---|
| Timepoint | ||||||||||
| 0 | 45.000000 | 45.000000 | 45.000000 | 45.000000 | 45.000000 | 45.000000 | 45.000000 | 45.000000 | 45.000000 | 45.000000 |
| 5 | 44.266086 | 46.503051 | 47.062001 | 47.389175 | 46.796098 | 47.125589 | 47.168130 | 43.944859 | 47.470830 | 46.851818 |
| 10 | 43.084291 | 48.285125 | 49.403909 | 49.582269 | 48.694210 | 49.423329 | 48.938560 | 42.531957 | 49.335368 | 48.689881 |
| 15 | 42.064317 | 50.094055 | 51.296397 | 52.399974 | 50.933018 | 51.359742 | 50.891769 | 41.495061 | 51.448025 | 50.779059 |
| 20 | 40.716325 | 52.157049 | 53.197691 | 54.920935 | 53.644087 | 54.364417 | 53.127384 | 40.238325 | 53.970080 | 53.170334 |
| 25 | 39.939528 | 54.287674 | 55.715252 | 57.678982 | 56.731968 | 57.482574 | 55.462490 | 38.974300 | 56.172821 | 55.432935 |
| 30 | 38.769339 | 56.769517 | 58.299397 | 60.994507 | 59.559509 | 59.809063 | 58.122548 | 38.703137 | 59.870528 | 57.713531 |
| 35 | 37.816839 | 58.827548 | 60.742461 | 63.371686 | 62.685087 | 62.420615 | 60.103457 | 37.451996 | 62.432021 | 60.089372 |
| 40 | 36.958001 | 61.467895 | 63.162824 | 66.068580 | 65.600754 | 65.052675 | 63.045537 | 36.574081 | 65.356386 | 62.916692 |
| 45 | 36.236114 | 64.132421 | 65.755562 | 70.662958 | 69.265506 | 68.084082 | 66.258529 | 34.955595 | 68.438310 | 65.960888 |
#This is where I set numerous counters and data that I will use for all charts and data, such as
# set the x_axis, colors, makers, and xlim for the various scatter charts, and a list and number of drugs
x_axis = np.arange(0, 50, 5)
colors = ['red','blue','green','yellow','tan','pink','violet','skyblue','orange','aqua']
markers = ['o','v','^','>','<','1','2','3','4','8']
plt.xlim(0,45)
drugs=dataset["Drug"].unique()
review_drugs=["Capomulin","Infubinol","Ketapril","Placebo"]
review_count=[0,1,2,3]
count = np.arange(0,len(drugs))
plt.style.use('seaborn-darkgrid')# First Chart of drug impact on tumors
plt.style.use('seaborn-darkgrid')
plt.title("Tumor Response to Treatment")
plt.xlabel("Time (Days)")
plt.ylabel("Tumor Volume in (mm3)")
plt.grid(alpha = 0.5)
for i in review_count:
chart_data = stats.sem(tumor_table[review_drugs[i]])
plt.errorbar(x_axis, tumor_table[review_drugs[i]], yerr = chart_data, marker= markers[i], color= colors[i], alpha = 0.5, label = review_drugs[i])
plt.legend(bbox_to_anchor=(1.05,1),loc= 2, borderaxespad = 0.)
plt.show()
# Chart of all drugs impact on tumors
plt.style.use('seaborn-darkgrid')
plt.title("Tumor Response to Treatment")
plt.xlabel("Time (Days)")
plt.ylabel("Tumor Volume in (mm3)")
plt.grid(alpha = 0.5)
for i in count:
chart_data = stats.sem(tumor_table[drugs[i]])
plt.errorbar(x_axis, tumor_table[drugs[i]], yerr = chart_data, marker= markers[i], color= colors[i], alpha = 0.5, label = drugs[i])
plt.legend(bbox_to_anchor=(1.05,1),loc= 2, borderaxespad = 0.)
plt.show()
# Metastatic Sites over time per drug
metastatic=dataset[['Mouse ID', 'Drug', 'Timepoint', 'Metastatic Sites']]
metastatic_data=pd.pivot_table(metastatic, index= ["Drug", "Timepoint"] )
metastatic_data.head(10)
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Metastatic Sites | ||
|---|---|---|
| Drug | Timepoint | |
| Capomulin | 0 | 0.000000 |
| 5 | 0.160000 | |
| 10 | 0.320000 | |
| 15 | 0.375000 | |
| 20 | 0.652174 | |
| 25 | 0.818182 | |
| 30 | 1.090909 | |
| 35 | 1.181818 | |
| 40 | 1.380952 | |
| 45 | 1.476190 |
#table of Metastatic results of all drugs over timepoints
metastatic_table=metastatic.pivot_table( index= "Timepoint", columns = "Drug", values = "Metastatic Sites")
metastatic_table
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Drug | Capomulin | Ceftamin | Infubinol | Ketapril | Naftisol | Placebo | Propriva | Ramicane | Stelasyn | Zoniferol |
|---|---|---|---|---|---|---|---|---|---|---|
| Timepoint | ||||||||||
| 0 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 |
| 5 | 0.160000 | 0.380952 | 0.280000 | 0.304348 | 0.260870 | 0.375000 | 0.347826 | 0.120000 | 0.260870 | 0.166667 |
| 10 | 0.320000 | 0.600000 | 0.666667 | 0.590909 | 0.523810 | 0.833333 | 0.619048 | 0.250000 | 0.523810 | 0.500000 |
| 15 | 0.375000 | 0.789474 | 0.904762 | 0.842105 | 0.857143 | 1.250000 | 0.800000 | 0.333333 | 0.809524 | 0.809524 |
| 20 | 0.652174 | 1.111111 | 1.050000 | 1.210526 | 1.150000 | 1.526316 | 1.000000 | 0.347826 | 0.947368 | 1.294118 |
| 25 | 0.818182 | 1.500000 | 1.277778 | 1.631579 | 1.500000 | 1.941176 | 1.384615 | 0.652174 | 1.166667 | 1.687500 |
| 30 | 1.090909 | 1.937500 | 1.588235 | 2.055556 | 2.066667 | 2.266667 | 1.666667 | 0.782609 | 1.411765 | 1.933333 |
| 35 | 1.181818 | 2.071429 | 1.666667 | 2.294118 | 2.266667 | 2.642857 | 2.333333 | 0.952381 | 1.533333 | 2.285714 |
| 40 | 1.380952 | 2.357143 | 2.100000 | 2.733333 | 2.466667 | 3.166667 | 2.777778 | 1.100000 | 1.583333 | 2.785714 |
| 45 | 1.476190 | 2.692308 | 2.111111 | 3.363636 | 2.538462 | 3.272727 | 2.571429 | 1.250000 | 1.727273 | 3.071429 |
#Chart of Metastatic results for review drugs
plt.style.use('seaborn-darkgrid')
plt.title("Metastatic Spread during Treatment")
plt.xlabel("Treatment Period (Days)")
plt.ylabel("Metastatic Sites")
plt.grid(alpha = 0.5)
for i in review_count:
chart_data = stats.sem(metastatic_table[review_drugs[i]])
plt.errorbar(x_axis, metastatic_table[review_drugs[i]], yerr = chart_data, marker= markers[i], color= colors[i], alpha = 0.5, label = review_drugs[i])
plt.legend(bbox_to_anchor=(1.05,1),loc= 2, borderaxespad = 0.)
plt.show()
#Chart of Metastatic results for all drugs
plt.style.use('seaborn-darkgrid')
plt.title("Metastatic Spread during Treatment")
plt.xlabel("Treatment Period (Days)")
plt.ylabel("Metastatic Sites")
plt.grid(alpha = 0.5)
for i in count:
chart_data = stats.sem(metastatic_table[drugs[i]])
plt.errorbar(x_axis, metastatic_table[drugs[i]], yerr = chart_data, marker= markers[i], color= colors[i], alpha = 0.5, label = drugs[i])
plt.legend(bbox_to_anchor=(1.05,1),loc= 2, borderaxespad = 0.)
plt.show() 
#subject mouse survival during the study
mice=dataset[['Mouse ID', 'Drug', 'Timepoint']]
mouse_count = mice.groupby(["Drug", "Timepoint"]).count()["Mouse ID"].rename("Mouse Count")
rodent_df=pd.DataFrame(mouse_count)
rodent_df.head(10)
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Mouse Count | ||
|---|---|---|
| Drug | Timepoint | |
| Capomulin | 0 | 25 |
| 5 | 25 | |
| 10 | 25 | |
| 15 | 24 | |
| 20 | 23 | |
| 25 | 22 | |
| 30 | 22 | |
| 35 | 22 | |
| 40 | 21 | |
| 45 | 21 |
mouse_table=rodent_df.pivot_table( index= "Timepoint", columns = "Drug", values = "Mouse Count")
mouse_table
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Drug | Capomulin | Ceftamin | Infubinol | Ketapril | Naftisol | Placebo | Propriva | Ramicane | Stelasyn | Zoniferol |
|---|---|---|---|---|---|---|---|---|---|---|
| Timepoint | ||||||||||
| 0 | 25 | 25 | 25 | 25 | 25 | 25 | 24 | 25 | 24 | 25 |
| 5 | 25 | 21 | 25 | 23 | 23 | 24 | 23 | 25 | 23 | 24 |
| 10 | 25 | 20 | 21 | 22 | 21 | 24 | 21 | 24 | 21 | 22 |
| 15 | 24 | 19 | 21 | 19 | 21 | 20 | 15 | 24 | 21 | 21 |
| 20 | 23 | 18 | 20 | 19 | 20 | 19 | 15 | 23 | 19 | 17 |
| 25 | 22 | 18 | 18 | 19 | 18 | 17 | 13 | 23 | 18 | 16 |
| 30 | 22 | 16 | 17 | 18 | 15 | 15 | 12 | 23 | 17 | 15 |
| 35 | 22 | 14 | 12 | 17 | 15 | 14 | 9 | 21 | 15 | 14 |
| 40 | 21 | 14 | 10 | 15 | 15 | 12 | 9 | 20 | 12 | 14 |
| 45 | 21 | 13 | 9 | 11 | 13 | 11 | 7 | 20 | 11 | 14 |
#chart of mouse survival rates per review drug
plt.style.use('seaborn-darkgrid')
plt.title("Mouse Survival During Treatment")
plt.xlabel("Time (Days)")
plt.ylabel("Surviving Mice")
plt.grid(alpha = 0.5)
for i in review_count:
chart_data = stats.sem(mouse_table[review_drugs[i]])
plt.errorbar(x_axis, mouse_table[review_drugs[i]], yerr = chart_data, marker= markers[i], color= colors[i], alpha = 0.5, label = review_drugs[i])
plt.legend(bbox_to_anchor=(1.05,1),loc= 2, borderaxespad = 0.)
plt.show()
#chart of mouse survival rates per all drugs
plt.style.use('seaborn-darkgrid')
plt.title("Mouse Survival During Treatment")
plt.xlabel("Time (Days)")
plt.ylabel("Surviving Mice")
plt.grid(alpha = 0.5)
for i in count:
chart_data = stats.sem(mouse_table[drugs[i]])
plt.errorbar(x_axis, mouse_table[drugs[i]], yerr = chart_data, marker= markers[i], color= colors[i], alpha = 0.5, label = drugs[i])
plt.legend(bbox_to_anchor=(1.05,1),loc= 2, borderaxespad = 0.)
plt.show() 
#rate of change for each drug
drugdelta=tumor_table.pct_change(periods=9).dropna()
percent_change = round(drugdelta * 100, 2)
percent_change
<style>
.dataframe thead tr:only-child th {
text-align: right;
}
</style>
.dataframe thead th {
text-align: left;
}
.dataframe tbody tr th {
vertical-align: top;
}
| Drug | Capomulin | Ceftamin | Infubinol | Ketapril | Naftisol | Placebo | Propriva | Ramicane | Stelasyn | Zoniferol |
|---|---|---|---|---|---|---|---|---|---|---|
| Timepoint | ||||||||||
| 45 | -19.48 | 42.52 | 46.12 | 57.03 | 53.92 | 51.3 | 47.24 | -22.32 | 52.09 | 46.58 |
# create the bar chart of tumor shrinkage in reviewed drugs
fig, ax = plt.subplots()
plt.style.use('seaborn-darkgrid')
performance = {}
for x in review_count:
performance[review_drugs[x]] = float(percent_change[review_drugs[x]])
x_axis = np.arange(0, len(review_drugs) )
tick_locations = []
for x in x_axis:
tick_locations.append(x + 0.4)
plt.xlim(-0.25, len(review_drugs))
plt.ylim(min(performance.values()) - 5, max(performance.values()) + 5)
plt.title("Tumor Volume Change Over 45 Day Treatment")
plt.ylabel("% Change in Tumor Size")
plt.axhline(y= 0, c='black')
bar_colors = pd.Series(list(performance.values()))
bar_colors = bar_colors > 0
bar_colors = bar_colors.map({True: "Red", False: "Green"})
plt.xticks(tick_locations, performance)
bar_chart= plt.bar(x_axis, performance.values(), color =bar_colors, alpha=0.75, align="edge")
def autolabels(rects):
for rect in rects:
h=rect.get_height()
ax.text(rect.get_x() + rect.get_width()/2., .8*h, '-%.2f%%' % float(h), ha='center', va='bottom', fontsize=10)
# label the bars
autolabels(bar_chart)
plt.show()
# create the bar chart of tumor shrinkage in all drugs
plt.style.use('seaborn-darkgrid')
fig, ax = plt.subplots()
performance = {}
for x in count:
performance[drugs[x]] = float(percent_change[drugs[x]])
x_axis = np.arange(0, len(drugs))
tick_locations = []
for x in x_axis:
tick_locations.append(x + 0.4)
plt.xlim(-0.25, len(drugs))
plt.ylim(min(performance.values()) - 5, max(performance.values()) + 5)
plt.title("Tumor Volume Change Over 45 Day Treatment")
plt.ylabel("% Change in Tumor Size")
plt.axhline(y= 0, c='black')
bar_colors = pd.Series(list(performance.values()))
bar_colors = bar_colors > 0
bar_colors = bar_colors.map({True: "Red", False: "Green"})
plt.xticks(tick_locations, performance, rotation="vertical" )
all_bar= plt.bar(x_axis, performance.values(), color =bar_colors, alpha=0.75, align="edge")
def autolabels(rects):
for rect in rects:
h=rect.get_height()
ax.text(rect.get_x() + rect.get_width()/2., .8*h, '%.0f%%' % float(h), ha='center', va='bottom', fontsize=10)
autolabels(all_bar)
plt.show()