Visualizing RBC Demographic Variables

Examining range, minimum and maximum values via boxplots

Examining distribution and frequency via histograms

import warnings
import glob

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

import numpy as np

# Define some general settings
pd.DataFrame.iteritems = pd.DataFrame.items
warnings.filterwarnings('ignore')
sns.set_theme(style='ticks')

# Let's read in our data
list_of_demographic_tsvs = sorted(glob.glob("*participants*tsv"))
print (list_of_demographic_tsvs)

['study-BHRC_desc-participants.tsv', 'study-CCNP_desc-participants.tsv', 'study-HBN_desc-participants.tsv', 'study-NKI_desc-participants.tsv', 'study-PNC_desc-participants.tsv']

Examining available variables:

dfs = [pd.read_table(f) for f in list_of_demographic_tsvs]
df = pd.concat(dfs, axis=0)
df = df.reset_index(drop=True)
print(df.columns.tolist())

['participant_id', 'session_id', 'study', 'study_site', 'wave', 'age', 'sex', 'race', 'ethnicity', 'bmi', 'handedness', 'participant_education', 'parent_1_education', 'parent_2_education', 'p_factor_mcelroy_harmonized_all_samples', 'internalizing_mcelroy_harmonized_all_samples', 'externalizing_mcelroy_harmonized_all_samples', 'attention_mcelroy_harmonized_all_samples', 'cubids_acquisition_group', 'session']

Variables of interest here are the age, sex, race, ethnicity, bmi, handedness, participant educatoin, parental education and p-factors.

Cleaning the dataset to ensure we only have the necessary information:

df = df[
    [
        'participant_id',
        'study',
        'age',
        'sex',
        'race',
        'ethnicity',
        'bmi',
        'handedness',
        'participant_education',
        'parent_1_education',
        'parent_2_education',
        'p_factor_mcelroy_harmonized_all_samples',
        'internalizing_mcelroy_harmonized_all_samples',
        'externalizing_mcelroy_harmonized_all_samples',
        'attention_mcelroy_harmonized_all_samples',
    ]
]
# Rename "Colornest" study to "CCNP"
df['study'] = df['study'].replace({'Colornest': 'CCNP'})
pal = sns.cubehelix_palette(len(df['study'].unique()), rot=-0.5, light=0.9)

Visualizing the variables

1. Age

Range

fig, ax = plt.subplots(figsize=(7, 6), dpi=400)

sns.boxplot(
    data=df,
    x='age',
    y='study',
    whis=[0, 100],
    width=.8,
    palette=pal,
    fliersize=8,
    ax=ax,
)

# Tweak the visual presentation
ax.set_ylabel('')
ax.set_title('RBC Age Range', fontsize=12)
ax.set_xlabel('Ages', fontsize=10)
ax.set_ylabel('Study', fontsize=10)
ax.set_xticks([0, 10, 20, 30, 40, 50, 60, 70, 80, 90])
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
fig.show()

Distribution

# Initialize the FacetGrid object
g = sns.FacetGrid(df, row='study', hue='study', aspect=9.5, height=1.5, palette=pal)

# Draw the densities in a few steps
g.map(
    sns.kdeplot,
    'age',
    bw_adjust=.5,
    clip_on=False,
    fill=True,
    alpha=1,
    linewidth=1.5,
)
g.map(sns.kdeplot, 'age', clip_on=False, color='w', lw=2, bw_adjust=.5)

# passing color=None to refline() uses the hue mapping
g.map(plt.axhline, y=0, linewidth=2, linestyle='-', color=None, clip_on=False)

# Define and use a simple function to label the plot in axes coordinates
def label(x, color, label):
    ax = plt.gca()
    ax.text(
        0.96,
        0.1,
        label,
        color='black',
        size=20,
        ha='left',
        va='center',
        transform=ax.transAxes,
    )

g.map(label, 'age')

# Set the subplots to overlap
g.fig.subplots_adjust(hspace=0)

# Remove axes details that don't play well with overlap
g.set(title='', yticks=[], ylabel='', xlabel='')
g.despine(bottom=True, left=True)
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
g.fig.suptitle('RBC Age Distribution', fontsize=24)
g.fig.supylabel('Study', fontsize=22)
g.fig.supxlabel('Age', fontsize=22)
plt.xticks(fontsize=15)
plt.show()

2. Sex

Distribution

fig, ax = plt.subplots(figsize=(4, 4), dpi=400)
df['sex'] = pd.Categorical(df['sex'], ['Female','Male','Other'])
x = sns.histplot(
    df,
    x='sex',
    hue='study',
    multiple='stack',
    palette=pal,
    ax=ax,
)
sns.despine(trim=False)
ax.set(xlabel='Sex', ylabel='Frequency', title='RBC Sex Distribution')
legend = ax.get_legend()
legend.get_title().set_fontsize(8)
for text in legend.texts:
    text.set_fontsize(6)

fig.show()

3. Race

Distribution

fig, ax = plt.subplots(figsize=(4, 4), dpi=400)
df['race'] = pd.Categorical(df['race'], ['Asian','Black','White','Other'])

x = sns.histplot(
    df,
    x='race',
    hue='study',
    multiple='stack',
    palette=pal,
    ax=ax,
)
sns.despine(trim=False)
ax.set(xlabel='Race', ylabel='Frequency', title='RBC Race Distribution')
legend = ax.get_legend()
legend.get_title().set_fontsize(8)
for text in legend.texts:
    text.set_fontsize(6)

fig.show()

4. Ethnicity

Distribution

fig, ax = plt.subplots(figsize=(4, 4), dpi=400)
x = sns.histplot(
    df,
    x='ethnicity',
    hue='study',
    multiple='stack',
    palette=pal,
    ax=ax,
)
sns.despine(trim=False)
ax.set(xlabel='Hispanic or Latino', ylabel='Frequency', title='RBC Ethnicity Distribution')

plt.xticks(np.arange(2), ['Yes', 'No'])

legend = ax.get_legend()
legend.get_title().set_fontsize(8)
for text in legend.texts:
    text.set_fontsize(6)

fig.show()

5. BMI

Range

fig, ax = plt.subplots(figsize=(7, 6), dpi=400)

sns.boxplot(
    data=df,
    x='bmi',
    y='study',
    whis=[0, 100],
    width=.8,
    palette=pal,
    fliersize=8,
    ax=ax,
)

# Tweak the visual presentation
ax.set_ylabel('')
ax.set_title('RBC BMI Range', fontsize=12)
ax.set_xlabel('BMIs', fontsize=10)
ax.set_ylabel('Study', fontsize=10)
ax.set_xticks([0, 10, 20, 30, 40, 50, 60, 70, 80, 90])
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
fig.show()

Distribution

# Initialize the FacetGrid object
g = sns.FacetGrid(df, row='study', hue='study', aspect=9.5, height=1.5, palette=pal)

# Draw the densities in a few steps
g.map(
    sns.kdeplot,
    'bmi',
    bw_adjust=.5,
    clip_on=False,
    fill=True,
    alpha=1,
    linewidth=1.5,
)
g.map(sns.kdeplot, 'bmi', clip_on=False, color='w', lw=2, bw_adjust=.5)

# passing color=None to refline() uses the hue mapping
g.map(plt.axhline, y=0, linewidth=2, linestyle='-', color=None, clip_on=False)

# Define and use a simple function to label the plot in axes coordinates
def label(x, color, label):
    ax = plt.gca()
    ax.text(
        0.96,
        0.1,
        label,
        color='black',
        size=20,
        ha='left',
        va='center',
        transform=ax.transAxes,
    )

g.map(label, 'bmi')

# Set the subplots to overlap
g.fig.subplots_adjust(hspace=0)

# Remove axes details that don't play well with overlap
g.set(title='', yticks=[], ylabel='', xlabel='')
g.despine(bottom=True, left=True)
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
g.fig.suptitle('RBC BMI Distribution', fontsize=24)
g.fig.supylabel('Study', fontsize=22)
g.fig.supxlabel('BMI', fontsize=22)
plt.xticks(fontsize=15)
plt.show()

6. Handedness

Distribution

fig, ax = plt.subplots(figsize=(4, 4), dpi=400)
x = sns.histplot(
    df,
    x='handedness',
    hue='study',
    multiple='stack',
    palette=pal[0:5],
    ax=ax,
)
sns.despine(trim=False)
ax.set(xlabel='Handedness', ylabel='Frequency', title='RBC Handedness Distribution')
legend = ax.get_legend()
legend.get_title().set_fontsize(8)
for text in legend.texts:
    text.set_fontsize(6)

fig.show()

7. Participant Education

Distribution

fig, ax = plt.subplots(figsize=(4, 4), dpi=400)
df['participant_education'] = pd.Categorical(df['participant_education'], ['Pre-Kindergarten',
                                                                           'Kindergarten',
                                                                           '1st Grade', '2nd Grade', 
                                                                           '3rd Grade', '4th Grade',
                                                                           '5th Grade', '6th Grade',
                                                                           '7th Grade', '8th Grade',
                                                                           '9th Grade', '10th Grade',
                                                                           '11th Grade', '12th Grade', 
                                                                           'High School Diploma',
                                                                           'Some College'])
x = sns.histplot(
    df,
    x='participant_education',
    hue='study',
    multiple='stack',
    palette=pal[0:5],
    ax=ax,
)
plt.xticks(rotation=90, ha='center')
sns.despine(trim=False)
ax.set(xlabel='Participant Education', ylabel='Frequency', title='RBC Participant Education Distribution')
legend = ax.get_legend()
legend.get_title().set_fontsize(8)
for text in legend.texts:
    text.set_fontsize(6)
sns.move_legend(ax, 'upper left', bbox_to_anchor=(1, 1))

fig.show()

8. Parent 1 Education

Distribution

fig, ax = plt.subplots(figsize=(4, 4), dpi=400)
df['parent_1_education'] = pd.Categorical(df['parent_1_education'], ['No/incomplete primary', 
                                                                     'Complete primary', 
                                                                     'Complete secondary', 
                                                                     'Complete tertiary'])
x = sns.histplot(
    df,
    x='parent_1_education',
    hue='study',
    multiple='stack',
    palette=pal[0:5],
    ax=ax,
)
plt.xticks(rotation=90, ha='center')
sns.despine(trim=False)
ax.set(xlabel='Parent 1 Education', ylabel='Frequency', title='RBC Parent 1 Distribution')
legend = ax.get_legend()
legend.get_title().set_fontsize(8)
for text in legend.texts:
    text.set_fontsize(6)

fig.show()

9. Parent 2 Education

Distribution

fig, ax = plt.subplots(figsize=(4, 4), dpi=400)
df['parent_2_education'] = pd.Categorical(df['parent_2_education'], ['No/incomplete primary', 
                                                                     'Complete primary', 
                                                                     'Complete secondary', 
                                                                     'Complete tertiary'])
x = sns.histplot(
    df,
    x='parent_2_education',
    hue='study',
    multiple='stack',
    palette=pal[0:5],
    ax=ax,
)
plt.xticks(rotation=90, ha='center')
sns.despine(trim=False)
ax.set(xlabel='Parent 2 Education', ylabel='Frequency', title='RBC Parent 2 Distribution')
legend = ax.get_legend()
legend.get_title().set_fontsize(8)
for text in legend.texts:
    text.set_fontsize(6)

fig.show()

10. Internalizing Factor

Range

fig, ax = plt.subplots(figsize=(7, 6), dpi=400)

sns.boxplot(
    data=df,
    x='internalizing_mcelroy_harmonized_all_samples',
    y='study',
    whis=[0, 100],
    width=.8,
    palette=pal,
    fliersize=8,
    ax=ax,
)

# Tweak the visual presentation
ax.set_ylabel('')
ax.set_title('RBC Internalizing Factor Range', fontsize=12)
ax.set_xlabel('Internalizing Factors', fontsize=10)
ax.set_ylabel('Study', fontsize=10)
ax.set_xticks([-3, -2, -1, 0, 1, 2, 3])
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
fig.show()

Distribution

# Initialize the FacetGrid object
g = sns.FacetGrid(df, row='study', hue='study', aspect=9.5, height=1.5, palette=pal)

# Draw the densities in a few steps
g.map(
    sns.kdeplot,
    'internalizing_mcelroy_harmonized_all_samples',
    bw_adjust=.5,
    clip_on=False,
    fill=True,
    alpha=1,
    linewidth=1.5,
)
g.map(sns.kdeplot, 'internalizing_mcelroy_harmonized_all_samples', clip_on=False, color='w', lw=2, bw_adjust=.5)

# passing color=None to refline() uses the hue mapping
g.map(plt.axhline, y=0, linewidth=2, linestyle='-', color=None, clip_on=False)

# Define and use a simple function to label the plot in axes coordinates
def label(x, color, label):
    ax = plt.gca()
    ax.text(
        0.96,
        0.1,
        label,
        color='black',
        size=20,
        ha='left',
        va='center',
        transform=ax.transAxes,
    )

g.map(label, 'internalizing_mcelroy_harmonized_all_samples')

# Set the subplots to overlap
g.fig.subplots_adjust(hspace=0)

# Remove axes details that don't play well with overlap
g.set(title='', yticks=[], ylabel='', xlabel='')
g.despine(bottom=True, left=True)
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
g.fig.suptitle('RBC Internalizing Factor Distribution', fontsize=24)
g.fig.supylabel('Study', fontsize=22)
g.fig.supxlabel('Internalizing Factors', fontsize=22)
plt.xticks(fontsize=15)
plt.show()

11. Externalizing Factor

Range

fig, ax = plt.subplots(figsize=(7, 6), dpi=400)

sns.boxplot(
    data=df,
    x='externalizing_mcelroy_harmonized_all_samples',
    y='study',
    whis=[0, 100],
    width=.8,
    palette=pal,
    fliersize=8,
    ax=ax,
)

# Tweak the visual presentation
ax.set_ylabel('')
ax.set_title('RBC Externalizing Factor Range', fontsize=12)
ax.set_xlabel('Externalizing Factor', fontsize=10)
ax.set_ylabel('Study', fontsize=10)
ax.set_xticks([-3, -2, -1, 0, 1, 2, 3])
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
fig.show()

Distribution

# Initialize the FacetGrid object
g = sns.FacetGrid(df, row='study', hue='study', aspect=9.5, height=1.5, palette=pal)

# Draw the densities in a few steps
g.map(
    sns.kdeplot,
    'externalizing_mcelroy_harmonized_all_samples',
    bw_adjust=.5,
    clip_on=False,
    fill=True,
    alpha=1,
    linewidth=1.5,
)
g.map(sns.kdeplot, 'externalizing_mcelroy_harmonized_all_samples', clip_on=False, color='w', lw=2, bw_adjust=.5)

# passing color=None to refline() uses the hue mapping
g.map(plt.axhline, y=0, linewidth=2, linestyle='-', color=None, clip_on=False)

# Define and use a simple function to label the plot in axes coordinates
def label(x, color, label):
    ax = plt.gca()
    ax.text(
        0.96,
        0.1,
        label,
        color='black',
        size=20,
        ha='left',
        va='center',
        transform=ax.transAxes,
    )

g.map(label, 'externalizing_mcelroy_harmonized_all_samples')

# Set the subplots to overlap
g.fig.subplots_adjust(hspace=0)

# Remove axes details that don't play well with overlap
g.set(title='', yticks=[], ylabel='', xlabel='')
g.despine(bottom=True, left=True)
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
g.fig.suptitle('RBC Externalizing Factor Distribution', fontsize=24)
g.fig.supylabel('Study', fontsize=22)
g.fig.supxlabel('Externalizing Factors', fontsize=22)
plt.xticks(fontsize=15)
plt.show()

12. Attention Factor

Range

fig, ax = plt.subplots(figsize=(7, 6), dpi=400)

sns.boxplot(
    data=df,
    x='attention_mcelroy_harmonized_all_samples',
    y='study',
    whis=[0, 100],
    width=.8,
    palette=pal,
    fliersize=8,
    ax=ax,
)

# Tweak the visual presentation
ax.set_ylabel('')
ax.set_title('RBC Attention Factor Range', fontsize=12)
ax.set_xlabel('Attention Factors', fontsize=10)
ax.set_ylabel('Study', fontsize=10)
ax.set_xticks([-3, -2, -1, 0, 1, 2, 3])
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
fig.show()

Distribution

# Initialize the FacetGrid object
g = sns.FacetGrid(df, row='study', hue='study', aspect=9.5, height=1.5, palette=pal)

# Draw the densities in a few steps
g.map(
    sns.kdeplot,
    'attention_mcelroy_harmonized_all_samples',
    bw_adjust=.5,
    clip_on=False,
    fill=True,
    alpha=1,
    linewidth=1.5,
)
g.map(sns.kdeplot, 'attention_mcelroy_harmonized_all_samples', clip_on=False, color='w', lw=2, bw_adjust=.5)

# passing color=None to refline() uses the hue mapping
g.map(plt.axhline, y=0, linewidth=2, linestyle='-', color=None, clip_on=False)

# Define and use a simple function to label the plot in axes coordinates
def label(x, color, label):
    ax = plt.gca()
    ax.text(
        0.96,
        0.1,
        label,
        color='black',
        size=20,
        ha='left',
        va='center',
        transform=ax.transAxes,
    )

g.map(label, 'attention_mcelroy_harmonized_all_samples')

# Set the subplots to overlap
g.fig.subplots_adjust(hspace=0)

# Remove axes details that don't play well with overlap
g.set(title='', yticks=[], ylabel='', xlabel='')
g.despine(bottom=True, left=True)
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
g.fig.suptitle('RBC Attention Factor Distribution', fontsize=24)
g.fig.supylabel('Study', fontsize=22)
g.fig.supxlabel('Attention Factors', fontsize=22)
plt.xticks(fontsize=15)
plt.show()

13. P-Factor

Range

fig, ax = plt.subplots(figsize=(7, 6), dpi=400)

sns.boxplot(
    data=df,
    x='p_factor_mcelroy_harmonized_all_samples',
    y='study',
    whis=[0, 100],
    width=.8,
    palette=pal,
    fliersize=8,
    ax=ax,
)

# Tweak the visual presentation
ax.set_ylabel('')
ax.set_title('RBC P-Factor Range', fontsize=12)
ax.set_xlabel('P-Factors', fontsize=10)
ax.set_ylabel('Study', fontsize=10)
ax.set_xticks([-3, -2, -1, 0, 1, 2, 3])
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
fig.show()

Distribution

# Initialize the FacetGrid object
g = sns.FacetGrid(df, row='study', hue='study', aspect=9.5, height=1.5, palette=pal)

# Draw the densities in a few steps
g.map(
    sns.kdeplot,
    'p_factor_mcelroy_harmonized_all_samples',
    bw_adjust=.5,
    clip_on=False,
    fill=True,
    alpha=1,
    linewidth=1.5,
)
g.map(sns.kdeplot, 'p_factor_mcelroy_harmonized_all_samples', clip_on=False, color='w', lw=2, bw_adjust=.5)

# passing color=None to refline() uses the hue mapping
g.map(plt.axhline, y=0, linewidth=2, linestyle='-', color=None, clip_on=False)

# Define and use a simple function to label the plot in axes coordinates
def label(x, color, label):
    ax = plt.gca()
    ax.text(
        0.96,
        0.1,
        label,
        color='black',
        size=20,
        ha='left',
        va='center',
        transform=ax.transAxes,
    )

g.map(label, 'p_factor_mcelroy_harmonized_all_samples')

# Set the subplots to overlap
g.fig.subplots_adjust(hspace=0)

# Remove axes details that don't play well with overlap
g.set(title='', yticks=[], ylabel='', xlabel='')
g.despine(bottom=True, left=True)
sns.despine(ax=ax, trim=False)
plt.tick_params(labelsize=8)
g.fig.suptitle('RBC P-Factor Distribution', fontsize=24)
g.fig.supylabel('Study', fontsize=22)
g.fig.supxlabel('P-Factors', fontsize=22)
plt.xticks(fontsize=15)
plt.show()