plotGraphs.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Jul 18 12:01:35 2018

@author: roshan
"""
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from collections import Counter
import seaborn as sns
from copy import deepcopy


def plotGraph(df, cat, plt_type='bar', 
              dropna=False, figsize=(6,4), title=None, ax=None,
              bins=10, print_bar_count=True): 
    '''
    Plots a graph of type defined in 'plt_type' on the data in the column given by 'cat'.
    Arg::
    plt_type: can be one of 'hist+density', 'hist', 'pie', 'bar', 'barh'  
    '''   
    sns.set(color_codes=False)
    
    if ax is None:
        fig, ax = plt.subplots(figsize=figsize)
    #set title
    if(title):
        ax.set_title(title)
    else:
        description = cat.split(" Index")[0]
        ax.set_title(description)
    
    data = df[cat]
    # if nans in the df then replace
    if(dropna): 
        data = data.dropna()
        data = data[(data!='NaN') or (data!="missing")]
    
    # histogram and density plots
    if (plt_type == 'hist+density'):
        data.plot.hist(grid = True, alpha=0.7, density=True, normed=True, bins=bins, ax=ax)
        data.plot.density(legend=True)
        ax.set_ylabel("density of subjects")
        
    # only histogram plots
    elif (plt_type == 'hist'):
        data.astype(float).plot.hist(grid=True, alpha=0.7, bins=bins, ax=ax)
        ax.set_ylabel("number of subjects")

         # pie-chart plots
    elif (plt_type == 'pie'):            
        val = dict(data.astype('str').value_counts())
        ax.pie(list(val.values()), labels = list(val.keys()), autopct='%1.1f%%', shadow=True, startangle=270) 
        ax.axis('equal')
        
    # bar plot
    elif (plt_type == 'bar'): 
        data.value_counts(dropna=dropna).sort_index().plot.bar(sort_columns=True, grid=True, rot=75, ax=ax,  width=0.9)
        if print_bar_count:
            x_texts = list(data.value_counts().sort_index())
            for i, x in enumerate(x_texts):
                ax.text( i , x+(x/100)+1 , str(x))
        ax.set_ylabel("number of subjects")
        
    # barh plot
    elif (plt_type == 'barh'): 
        data.value_counts(dropna=dropna).sort_index().plot.barh(ax=ax, width=0.9)
        
        if print_bar_count:
            y_texts = list(data.value_counts().sort_index())
            for i, y in enumerate(y_texts):
                ax.text(y+(y/100), i, str(y))
        ax.set_xlabel("number of subjects")
    
    else:
        raise ValueError("Invalid value for plt_type. It can be one of 'hist+density', 'hist', 'pie', 'bar', barh") 
        
        
        
######################## UKBiobank: ICD related plotting functions ################################################################

ICD10map = {
    "A00-B99":"Certain infectious and parasitic diseases",
    "C00-D48":"Neoplasms",
    "D50-D89":"Diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism",
    "E00-E90":"Endocrine, nutritional and metabolic diseases",
    "F00-F99":"Mental and behavioural disorders",
    "G00-G99":"Diseases of the nervous system",
    "H00-H59":"Diseases of the eye and adnexa",
    "H60-H95":"Diseases of the ear and mastoid process",
    "I00-I99":"Diseases of the circulatory system",
    "J00-J99":"Diseases of the respiratory system",
    "K00-K93":"Diseases of the digestive system",
    "L00-L99":"Diseases of the skin and subcutaneous tissue",
    "M00-M99":"Diseases of the musculoskeletal system and connective tissue",
    "N00-N99":"Diseases of the genitourinary system",
    "O00-O99":"Pregnancy, childbirth and the puerperium",
    "P00-P96":"Certain conditions originating in the perinatal period",
    "Q00-Q99":"Congenital malformations, deformations and chromosomal abnormalities",
    "R00-R99":"Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified",
    "S00-T98":"Injury, poisoning and certain other consequences of external causes",
    "V01-Y98":"External causes of morbidity and mortality",
    "Z00-Z99":"Factors influencing health status and contact with health services",
    "U00-U99":"Codes for special purposes" }

ICD10_G_map = {
    "G10-G14":"Inflammatory diseases of the central nervous",
    "G20-G26":"Systemic atrophies primarily affecting the central nervous",
    "G30-G32":"Other degenerative diseases of the nervous system",
    "G35-G37":"Demyelinating diseases of the central nervous system",
    "G40-G47":"Episodic and paroxysmal disorders",
    "G50-G59":"Nerve, nerve root and plexus disorders",
    "G60-G64":"Polyneuropathies and other disorders of the peripheral nervous system",
    "G70-G73":"Diseases of myoneural junction and muscle",
    "G80-G83":"Cerebral palsy and other paralytic syndromes",
    "G90-G99":"Other disorders of the nervous system" }

ICD10_F_map = {
    "F00-F09":"Organic, including symptomatic, mental disorders",
    "F10-F19":"Mental and behavioural disorders due to psychoactive substance use",
    "F20-F29":"Schizophrenia, schizotypal and delusional disorders",
    "F30-F39":"Mood [affective] disorders",
    "F40-F48":"Neurotic, stress-related and somatoform disorders",
    "F50-F59":"Behavioural syndromes associated with physiological disturbances and physical factors",
    "G60-G64":"Polyneuropathies and other disorders of the peripheral nervous system",
    "F60-F69":"Disorders of adult personality and behaviour",
    "F70-F79":"Mental retardation",
    "F80-F89":"Disorders of psychological development",
    "F90-F98":"Behavioural and emotional disorders with onset usually occurring in childhood and adolescence",
    "F99-F99":"Unspecified mental disorder"}

def plotIcd10Distribution(df, CategoryMap=lambda x:x[:3], 
                          threshold=0, exclude_cats=[],
                          pieChart=False, title=None, rot=60):
    '''
Maps the categories using the 'CategoryMap' function to parent categories like in the 'ICD10map' mentioned above
and then plots a bar chart and a pie chart of the distribution of these mapped categories
if 'threshold' is given, Categories which have a count lower than or equal to the threshold are dropped'''
    cat_counter = Counter()
    
    assert "Diagnoses - all ICD10" in df.columns, "this func only works on if df has a column 'Diagnoses - all ICD10' that contain list of ICD diagnosis codes per subject"
    for each_subject in df["Diagnoses - all ICD10"]:
        #get a list of mapped ICD parent categories. remove duplicate parent categories
        parent_cats = list(set([CategoryMap(each_cat) for each_cat in each_subject]))
        # count each category
        for each_parent_cat in parent_cats:
            cat_counter[each_parent_cat] += 1
            
    #remove categories whose counts are lower than the threshold given
    cat_counter_copy = deepcopy(cat_counter)
    for cat, count in cat_counter_copy.items():
        catname = 'none' if cat is None else cat.lower()
        if(count <= threshold) or np.any([exclude in catname for exclude in exclude_cats]):
            del cat_counter[cat]
    
    #ignore 'None' category if it exists    
    try:
        del cat_counter[None]
    except KeyError:
        pass
    
    # convert the counts to a pandas dataframe for easy plotting
    ICD10_count_df = pd.DataFrame.from_dict(cat_counter, 'index').sort_index()
    ICD10_count_df.columns = ["number of subjects"]
    
    #bar plot
    f, ax = plt.subplots(figsize=(10,8))
    ICD10_count_df.plot.bar(grid=True, rot=rot, ax=ax)
    txt_pos = max(ICD10_count_df["number of subjects"]) //100
    for i, x in enumerate(ICD10_count_df["number of subjects"]):
        ax.text( i , x + txt_pos, x)
    
    if(title):
        ax.set_title(title)
    
    #pie plot
    if(pieChart):
        f2, ax2 = plt.subplots(figsize=(10,8))
        ax2.pie(list(ICD10_count_df["number of subjects"]), labels = ICD10_count_df.index,
                autopct='%1.1f%%', shadow=True, startangle=90) 
        ax2.axis('equal')
    
    del(ICD10_count_df)
    
    return cat_counter


'''Takes a string denoting a ICD10 category ex. "S91" and returns a numeric value for comparison'''
def getICDVal(strVal):
    letter = ord(strVal[0])*100
    num = int(strVal[1:])
    return (letter + num)


'''Takes a string range of the ICD10 categories ex. "V01-Y98" and returns the range in numeric form for comparison'''
def getICDRange(strRange):
    l,u = strRange.split("-")
    return range(getICDVal(l), getICDVal(u)+1)

'''Given the numeric version of a ICD10 category, this func returns 
the parent category and the description of the category'''
def mapICDParents(category):
    # first check if the category is empty
    if(category is np.nan):
        return "NaN", "No data / Healthy"
    category = category[:3]
    for key in ICD10map.keys():
        if getICDVal(category) in getICDRange(key):
            return key
    # raise error if an invalid category is given
    raise ValueError("Invalid category - {}".format(category))
    
    
# Plot distribution of 'F00-F99' category 'Mental and behavioural disorders'
def mapFCategory(category):    
    category=  category[:3] 
    for key in ICD10_F_map.keys():
        if getICDVal(category) in getICDRange(key):
            return key
    return None