aiyanalizmathew / pymaceuticals Goto Github PK

# Dependencies
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from scipy.stats import sem
import os

# Setting the file path for clinical trial data file
filepath1=os.path.join("raw_data","clinicaltrial_data.csv") 
# Setting the file path for mouse drug data file
filepath2=os.path.join("raw_data","mouse_drug_data.csv")

# Reading the csv file
Clinical_data=pd.read_csv(filepath1)
Mouse_Drug_data=pd.read_csv(filepath2)

# Combining the data files to one single dataframe
Clinical_Combined_data=pd.merge(Clinical_data,Mouse_Drug_data,on="Mouse ID",how="left")

# Creating a pivot table for Clinical Tumor Response
Clinical_TumorResponse=pd.pivot_table(Clinical_Combined_data, values="Tumor Volume (mm3)",
                                      columns=["Drug"],index=["Timepoint"],aggfunc=np.mean)
Clinical_TumorResponse

# Droping the not needed drugs from the already existing pivot table
Clinical_TumorResponse_DF=Clinical_TumorResponse.drop(['Ceftamin','Naftisol','Propriva','Ramicane','Stelasyn','Zoniferol'],
                                                      axis=1)
Clinical_TumorResponse_DF

# Setting the lists of drugs, colors and markers respectively
drug=["Capomulin","Infubinol","Ketapril","Placebo"]
colors=["red","blue","green","black"]
markers=["o","^","s","D"]

# Intializing the plots
fig, ax = plt.subplots()

for item in drug:
    # Setting the x_axis data,y_axis data and standard deviation for the graph
    x_data=Clinical_TumorResponse_DF.index.values
    y_data=Clinical_TumorResponse_DF[item] 
    err=sem(Clinical_TumorResponse_DF[item])
    
    # Finding the index for colors and markers
    index=drug.index(item)
    # Setting up the plot
    ax.errorbar(x_data, y_data,err,linestyle="--",color=colors[index],
                marker=markers[index],capsize=3,capthick=1,markeredgecolor='black')
    # Setting the legend
    ax.legend(loc="upper left", fancybox=True,numpoints=2,edgecolor="black")

# Setting the grid   
plt.grid(linestyle='dotted')
# Setting the x_axis and y_axis limits
ax.set_xlim(0,45)
ax.set_ylim(0,80)
# Setting the title,x_axis and y_axis labels
ax.set_title("Tumor Response to Treatment", fontsize=16)
ax.set_xlabel("Time(Days)")
ax.set_ylabel("Tumor Volume (mm3)")

# Display the graph
plt.show()

# Creating a pivot table for Metastatic Response
Metastatic_Response=pd.pivot_table(Clinical_Combined_data, values="Metastatic Sites",
                                   columns=["Drug"],index=["Timepoint"],aggfunc=np.mean)
Metastatic_Response

# Droping the not needed drugs from the already existing pivot table
Metastatic_Response_DF=Metastatic_Response.drop(['Ceftamin','Naftisol','Propriva','Ramicane','Stelasyn','Zoniferol'], axis=1)

Metastatic_Response_DF

# Setting the lists of drugs, colors and markers respectively
drug=["Capomulin","Infubinol","Ketapril","Placebo"]
colors=["red","blue","green","black"]
markers=["o","^","s","D"]

# Intializing the plots
fig, ax = plt.subplots()

for item in drug:
    # Setting the x_axis data,y_axis data and standard deviation for the graph
    x_data=Metastatic_Response_DF.index.values
    y_data=Metastatic_Response_DF[item]
    err=sem(Metastatic_Response_DF[item])
    
    # Finding the index for colors and markers
    index=drug.index(item)
    
    # Setting up the plot
    ax.errorbar(x_data, y_data,err,linestyle=":",color=colors[index],
                marker=markers[index],capsize=3,capthick=1,markeredgecolor='black')
    # Setting the legend
    ax.legend(loc="upper left", fancybox=True,numpoints=2,edgecolor="black")

# Setting the grid     
plt.grid(linestyle='dotted')
# Setting the x_axis and y_axis limits
ax.set_xlim(0,45)
ax.set_ylim(0,4)
# Setting the title,x_axis and y_axis labels
ax.set_title("Metastatic Spread during Treatment", fontsize=16)
ax.set_xlabel("Treatment Duration (Days)")
ax.set_ylabel("Met. Sites")

# Display the graph
plt.show()

# Creating a pivot table for Survival Rates
Survival_Rates=pd.pivot_table(Clinical_Combined_data, values="Tumor Volume (mm3)",
                              columns=["Drug"],index=["Timepoint"],aggfunc='count')

Survival_Rates

# Droping the not needed drugs from the already existing pivot table
Survival_Rates_DF=Survival_Rates.drop(['Ceftamin','Naftisol','Propriva','Ramicane','Stelasyn','Zoniferol'], axis=1)

Survival_Rates_DF

# Setting the lists of drugs, colors and markers respectively
drug=["Capomulin","Infubinol","Ketapril","Placebo"]
colors=["red","blue","green","black"]
markers=["o","^","s","D"]

# Intializing the plots
fig, ax = plt.subplots()

for item in drug:
    # Setting the x_axis data and y_axis data for the graph
    x_data=Survival_Rates_DF.index.values
    y_data=(Survival_Rates_DF[item]*100)/ Survival_Rates_DF[item][0]
    
    # Finding the index for colors and markers
    index=drug.index(item)
    
    # Setting up the plot
    ax.plot(x_data,y_data,linestyle="--",color=colors[index],marker=markers[index],label=item,markeredgecolor='black')
    # Setting the legend
    ax.legend(loc="lower left", fancybox=True,numpoints=2,edgecolor="black")
    
# Setting the grid
ax.grid(linestyle='dotted')
# Setting the x_axis and y_axis limits
ax.set_xlim(0,45)
ax.set_ylim(0,102)
# Setting the title,x_axis and y_axis labels
ax.set_title("Survival during Treatment", fontsize=16)
ax.set_xlabel("Time (Days)")
ax.set_ylabel("Survival Rate(%)")

# Display the graph
plt.show()

# Summary of the analysis based on % Tumor Change over 45 days
Summary=round(((Clinical_TumorResponse_DF.iloc[9,:]-Clinical_TumorResponse_DF.iloc[0,:])*100/Clinical_TumorResponse_DF.iloc[0,:]),0).to_frame()
# Renaming the column
Summary=Summary.rename(columns={0:"% Tumor Change"})

Summary

# Setting the x_axis and y-axis value
x_axis=Summary.index.values
y_axis=Summary["% Tumor Change"]

# If value is positive then put True in the Summary else place False, for changing the color based on the value
Summary["Positive"]=Summary["% Tumor Change"]>0

# Intializing the plots
fig,ax=plt.subplots()
#Setting the plot and assigning the color based on the Positive or not value
bars=ax.bar(x_axis,y_axis,align="edge",width=1,linewidth=1,
            edgecolor=Summary.Positive.map({True: 'black', False: 'black'}),
            color=Summary.Positive.map({True: 'green', False: 'red'}))

# Setting the ticks for the bar graph
tick_locations = [value+0.5 for value in range(len(x_axis))]
ax.set_xticks(tick_locations)

# Setting the text label in the bar graph
# Assign the height based on the positive calue
height = Summary.Positive.map({True: 5, False: -5})
for bar in bars:
    ax.text(bar.get_x() + bar.get_width()/2., height[bars.index(bar)],
            str(int(Summary["% Tumor Change"][bars.index(bar)]))+"%",
                ha='center', va='bottom')


# Setting the x_axis limits
ax.set_xlim(0, len(x_axis))
# Setting a horizontal line at y=0
plt.hlines(0,0,len(x_axis))
# Setting the title of the graph
ax.set_title("Tumor Change over 45 Day Treatment", fontsize=16)
# Setting the y_axis label
ax.set_ylabel("% Tumor Volume Change")

# Setting the grid line
plt.grid(linestyle='dotted')

# Display the graph
plt.show()