-
Notifications
You must be signed in to change notification settings - Fork 0
/
version 2
257 lines (223 loc) · 11 KB
/
version 2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
import os
import streamlit as st
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import altair as alt
def generate_plots(combined_df, original_df, ppmtol):
try:
# Create the fdval dataframe by combining the ScanNum and MonoisotopicMass columns from the combined_df
fdval = combined_df[['ScanNum', 'MonoisotopicMass']]
# Convert the ScanNum column to integers
fdval['ScanNum'] = fdval['ScanNum'].astype(int)
# Create the tpindex1 logical index by checking if the values in the ScanNum column of fdval are present in the Scan column of original_df
tpindex1 = fdval['ScanNum'].isin(original_df['ScanNum'].dropna().replace([np.inf, -np.inf], np.nan).astype(int))
# Adjusted code to allow for ±1 Da error with user-defined ppmtol Da tolerance
tpindex2 = np.logical_or.reduce([
(fdval.iloc[:, 1].values[:, np.newaxis] >= original_df.iloc[:, 2].values - ppmtol / 10) &
(fdval.iloc[:, 1].values[:, np.newaxis] <= original_df.iloc[:, 2].values + ppmtol / 10),
(fdval.iloc[:, 1].values[:, np.newaxis] >= original_df.iloc[:, 2].values - (ppmtol + 1) / 10) &
(fdval.iloc[:, 1].values[:, np.newaxis] <= original_df.iloc[:, 2].values - (ppmtol - 1) / 10),
(fdval.iloc[:, 1].values[:, np.newaxis] >= original_df.iloc[:, 2].values + (ppmtol - 1) / 10) &
(fdval.iloc[:, 1].values[:, np.newaxis] <= original_df.iloc[:, 2].values + (ppmtol + 1) / 10)
])
# Create the tpindex logical index by combining tpindex1, tpindex2, and the DummyIndex column from combined_df with values of 0
tpindex = np.logical_and.reduce((
tpindex1.values.flatten(),
tpindex2.any(axis=1),
combined_df['TargetDecoyType'].values == 0
))
# Identify false positives and decoys
fpindex = np.logical_and.reduce((
np.logical_not(tpindex),
combined_df['TargetDecoyType'].values == 0
))
decoyindex = combined_df['TargetDecoyType'] > 0
# Identify true positives, false positives, and decoys
true_positives = combined_df.loc[tpindex]
false_positives = combined_df.loc[fpindex]
decoy_masses = combined_df.loc[decoyindex]
# Count the number of instances for each category
tp_count = len(true_positives)
fp_count = len(false_positives)
decoy_count = len(decoy_masses)
# Plotting the histograms, ECDF, and KDE
st.subheader('Histogram of Qscore')
fig_hist = plt.figure(figsize=(8, 6))
plt.hist(combined_df.loc[fpindex, 'Qscore'], bins=100, alpha=0.7, label='False Positive Masses', color='red',
edgecolor='grey')
plt.hist(combined_df.loc[tpindex, 'Qscore'], bins=100, alpha=0.6, label='True Positive Masses', color='green',
edgecolor='grey')
plt.hist(combined_df.loc[decoyindex, 'Qscore'], bins=100, alpha=0.4, label='Decoy Masses', color='blue',
edgecolor='grey')
plt.xlabel('Qscore')
plt.ylabel('Count')
plt.legend()
plt.grid(True)
st.pyplot(fig_hist)
st.subheader('Empirical Cumulative Distribution Function')
fig_ecdf = plt.figure(figsize=(10, 6))
sns.ecdfplot(data=combined_df.loc[tpindex, 'Qscore'], label='True Positive Masses', color='green')
sns.ecdfplot(data=combined_df.loc[fpindex, 'Qscore'], label='False Positive Masses', color='red')
sns.ecdfplot(data=combined_df.loc[decoyindex, 'Qscore'], label='Decoy Masses', color='blue')
plt.xlabel('Qscore')
plt.ylabel('ECDF')
plt.legend()
plt.grid(True)
st.pyplot(fig_ecdf)
def altair_scatter_plot(data):
# Define the color scheme for different types of points
color_scale = alt.Scale(domain=['True Positives', 'False Positives', 'Decoys'],
range=['green', 'red', 'blue'])
# Create the scatter plot
scatter_chart = alt.Chart(data).mark_circle(size=60).encode(
x='ScanNum',
y='MonoisotopicMass',
color=alt.Color('Type', scale=color_scale),
tooltip=['ScanNum', 'MonoisotopicMass', 'Type']
).interactive()
# Display the chart in Streamlit``
st.altair_chart(scatter_chart, use_container_width=True)
st.subheader('Kernel Density Estimate')
fig_kde = plt.figure(figsize=(10, 6))
sns.kdeplot(data=combined_df.loc[tpindex, 'Qscore'], label='True Positive Masses', shade=True, color='green')
sns.kdeplot(data=combined_df.loc[fpindex, 'Qscore'], label='False Positive Masses', shade=True, color='red')
sns.kdeplot(data=combined_df.loc[decoyindex, 'Qscore'], label='Decoy Masses', shade=True, color='blue')
plt.xlabel('Qscore')
plt.ylabel('Density')
plt.legend()
st.pyplot(fig_kde)
# Pie chart for true positives, false positives, and decoy masses
st.subheader('Pie Chart: True Positives, False Positives, Decoy Masses')
fig_pie, ax_pie = plt.subplots()
labels = ['True Positives', 'False Positives', 'Decoy Masses']
sizes = [tp_count, fp_count, decoy_count]
colors = ['green', 'red', 'blue']
ax_pie.pie(sizes, labels=labels, autopct='%1.1f%%', colors=colors, startangle=90)
ax_pie.axis('equal') # Equal aspect ratio ensures that pie is drawn as a circle.
st.pyplot(fig_pie)
########################
###########################
##############
#######
# Calculate the mean Qvalue for bins of Qscore2D for true and decoy data
bin_means_true = true_positives.groupby(pd.cut(true_positives['Qscore2D'], np.arange(0, 1.05, 0.05)))[
'Qvalue'].mean()
bin_means_decoy = false_positives.groupby(pd.cut(false_positives['Qscore2D'], np.arange(0, 1.05, 0.05)))[
'Qvalue'].mean()
# Create the plot
fig, ax = plt.subplots(figsize=(10, 6))
# Line plot for the mean Qvalues
plt.plot(bin_means_true.index.categories.mid, bin_means_true.values, color='green',
label='True FDR')
plt.plot(bin_means_decoy.index.categories.mid, bin_means_decoy.values, color='red', linestyle='dashed',
label='Estimated FDR')
# Labeling
plt.xlabel('Qscore2D')
plt.ylabel('Qvalue')
plt.title('Score Distribution of Estimated FDR vs. True FDR')
plt.legend()
plt.grid(True)
st.pyplot(fig)
##########
st.subheader('Distribution Plot: Different DecoyTypes')
fig_dist, ax_dist = plt.subplots(figsize=(10, 6))
sns.kdeplot(data=combined_df.loc[combined_df['TargetDecoyType'] == 1, 'Qscore'], label='Noise Decoys',
shade=True)
sns.kdeplot(data=combined_df.loc[combined_df['TargetDecoyType'] == 2, 'Qscore'], label='Isotope Decoys',
shade=True)
sns.kdeplot(data=combined_df.loc[combined_df['TargetDecoyType'] == 3, 'Qscore'], label='Charge Decoys',
shade=True)
plt.xlabel('Qscore')
plt.ylabel('Density')
plt.legend()
st.pyplot(fig_dist)
###############################################
def altair_histogram(data, title):
# Create the histogram
histogram = alt.Chart(data).mark_bar().encode(
alt.X('Qscore:Q', bin=True),
y='count()',
color='Type:N',
tooltip=['count()', 'mean(Qscore)']
).properties(
title=title,
width=600,
height=400
).interactive()
# Display the chart in Streamlit
st.altair_chart(histogram, use_container_width=True)
################################################
########################
# New Plot for Score Distribution testing 1
st.subheader('Score Distribution testing1 ')
fig, ax = plt.subplots(figsize=(10, 6))
# Filter for TargetDecoyType=0
target_data = combined_df[combined_df['TargetDecoyType'] == 0]
# Filter for TargetDecoyType>0 (Decoy Masses)
decoy_data = combined_df[combined_df['TargetDecoyType'] > 0]
# Plot for Target Masses
sns.scatterplot(x='Qscore2D', y='Qvalue', data=true_positives, ax=ax, color='green', label='True FDR')
# Plot for Decoy Masses
sns.scatterplot(x='Qscore2D', y='Qvalue', data=false_positives, ax=ax, color='red', label='Estimated FDR')
plt.xlabel('Qscore2D')
plt.ylabel('Qvalue')
plt.title('Score Distribution of Target vs. Decoy Masses')
plt.legend()
st.pyplot(fig)
except Exception as e:
st.error(f"An error occurred: {str(e)}")
except Exception as e:
st.error(f"An error occurred: {str(e)}")
# Sidebar for file uploads
# Function to load data from a given file path
def load_data(file_path):
if os.path.exists(file_path):
return pd.read_csv(file_path)
else:
return pd.DataFrame() # Return an empty DataFrame if the file doesn't exist
# Sidebar for file uploads
st.sidebar.header("File Uploads")
combined_file = st.sidebar.file_uploader("Upload Combined File (CSV)", type=["csv"])
original_file = st.sidebar.file_uploader("Upload Original File (CSV)", type=["csv"])
# Sidebar for other options
ppmtol = st.sidebar.slider("Set ppmtol", min_value=1, max_value=20, value=5)
# Title
st.title('FDR Estimation FLASH App 🚀')
# Subheader for ppmtol
st.subheader(f'Checkout distributions plots for different tolerance: {ppmtol} ppm Tolerance')
# Paths for the example files
combined_example_path = "/Users/ayeshaferoz/Downloads/Res35k,noise1e3,centroid/FLASHout/adder.csv"
original_example_path = "/Users/ayeshaferoz/Downloads/chosen.csv"
# Load example or uploaded data
if combined_file is not None:
combined_df = pd.read_csv(combined_file)
st.write('Input file from FLASHDeconv :', combined_df)
else:
combined_df = load_data(combined_example_path)
st.write('FLASHDeconv Example output File:', combined_df)
if original_file is not None:
original_df = pd.read_csv(original_file)
st.write('True Mass List:', original_df)
else:
original_df = load_data(original_example_path)
st.write('True Mass List:', original_df)
# Check if dataframes are not empty and then generate plots
if not combined_df.empty and not original_df.empty:
st.subheader('Generate Plots')
if st.button('Generate Plots'):
generate_plots(combined_df, original_df, ppmtol)
# Information Section
st.write('---') # Horizontal line to separate sections
st.header('Information')
# Contact Us
st.subheader('Contact Us')
st.markdown("""
- **Email**: [[email protected]](mailto:[email protected])
""")
# Publications
st.subheader('Publications')
st.markdown("""
- Jeong et al., 2020, Cell Systems 10, 213–218. February 26, 2020. A 2020 The Authors. Published by Elsevier Inc. [DOI](https://doi.org/10.1016/j.cels.2020.01.003)
""")