File size: 9,332 Bytes
5dc2016
62586c5
fc731db
 
8ed723b
 
 
 
 
 
28995ec
98ac1da
28995ec
275028e
8ed723b
 
 
734d022
4a37eb1
ddff90b
9f09f8c
e782c79
98ac1da
68a2646
a4dfd52
 
 
 
6792ef6
9f09f8c
 
221e51c
fc731db
e7caceb
6d2e57c
e7caceb
 
6d2e57c
e7caceb
4a37eb1
e7caceb
 
4a37eb1
 
e7caceb
6d2e57c
e7caceb
 
 
6d2e57c
a400b1e
 
 
 
 
 
 
 
 
3239353
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
01f6525
3239353
 
01f6525
3239353
01f6525
3239353
 
 
01f6525
3239353
 
01f6525
3239353
 
9f09f8c
c9b72ab
fb5e624
9f09f8c
 
f2852e3
38efeba
847adc5
a224dfa
0416a61
f2852e3
ddff90b
9f09f8c
cde5ff7
b102419
8ed723b
 
 
 
fc731db
9f09f8c
7780086
091df08
 
 
fc731db
9f09f8c
27bf394
 
 
04706b7
fc731db
0d9531e
b28ab8e
 
 
062e24e
f435314
e1cbd0e
b28ab8e
 
 
f435314
 
04706b7
b28ab8e
062e24e
 
 
 
 
fc731db
b424a32
0d9531e
9f09f8c
8f768aa
8ed723b
a8b6710
9f09f8c
fb31138
9f09f8c
a8b6710
 
fb31138
7ead1f4
6d2e57c
26498e1
28275dd
801aede
 
2b7ba7c
 
 
 
26498e1
2b7ba7c
 
 
 
 
 
 
 
 
26498e1
 
5fe6306
26498e1
2b7ba7c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import nltk
nltk.data.path.append("/home/user/app/nltk_data")
#nltk.download('stopwords')
#nltk.download('punkt')
from epi_pipeline import (
    streamlit_extraction,
    NER_Pipeline,
    GARD_Search,
    Classify_Pipeline
    )
import pandas as pd
#pd.set_option('display.max_colwidth', None)
import streamlit as st
st.set_page_config(layout="wide")
#import spacy
#import tensorflow as tf
#import pickle
import re
import plotly.graph_objects as go

#### LOGO ####
st.markdown('''<img src="https://huggingface.co/spaces/ncats/EpiPipeline4RD/raw/main/ncats.png" alt="National Center for Advancing Translational Sciences Logo" width=600>''',unsafe_allow_html=True)
st.markdown("")
st.markdown('''<img src="https://huggingface.co/spaces/ncats/EpiPipeline4RD/resolve/main/Logo_GARD_fullres.png" alt="NIH Genetic and Rare Diseases Information Center Logo"  width=400>''',unsafe_allow_html=True)

#st.markdown('''<img src="https://huggingface.co/spaces/ncats/EpiPipeline4GARD/raw/main/ncats.svg" alt="National Center for Advancing Translational Sciences Logo" width=800>''',unsafe_allow_html=True)
#st.markdown("")
#st.markdown('''<img src="https://huggingface.co/spaces/ncats/EpiPipeline4GARD/resolve/main/Logo_GARD_fullres.png" alt="NIH Genetic and Rare Diseases Information Center Logo" width=800>''',unsafe_allow_html=True)
#st.markdown("![National Center for Advancing Translational Sciences (NCATS) Logo](https://huggingface.co/spaces/ncats/EpiPipeline4GARD/resolve/main/NCATS_logo.png)")

#### TITLE ####
st.title("Epidemiological Information Extraction Pipeline for Rare Diseases")
#st.subheader("National Center for Advancing Translational Sciences (NIH/NCATS)")

#### CHANGE SIDEBAR WIDTH ###
st.markdown(
    """
    <style>
    [data-testid="stSidebar"][aria-expanded="true"] > div:first-child {
        width: 250px;
    }
    [data-testid="stSidebar"][aria-expanded="false"] > div:first-child {
        width: 250px;
        margin-left: -350px;
    }
    </style>
    """,
    unsafe_allow_html=True,
)

### Remove 'Made with Streamlit' Footer
st.markdown("""
            <style>
            #MainMenu {visibility: hidden;}
            footer {visibility: hidden;}
            header {visibility: hidden;}
            </style>
            """, unsafe_allow_html=True)

### ADD NEW FOOTER
footer="""<style>
a:link , a:visited{
color: blue;
background-color: transparent;
text-decoration: underline;
}

a:hover,  a:active {
color: red;
background-color: transparent;
text-decoration: underline;
}

.footer {
position: fixed;
left: 10;
bottom: 0;
width: 100%;
background-color: transparent;
color: black;
text-align: left;
}
</style>
<div class="footer">
<p>Developed by <a style='display: block; text-align: center;' href="https://github.com/wzkariampuzha" target="_blank">William Kariampuzha at NIH/NCATS</a></p>
</div>
"""
#st.markdown(footer,unsafe_allow_html=True)


#### DESCRIPTION ####
st.markdown("This application was built by the [National Center for Advancing Translational Sciences (NCATS)](https://ncats.nih.gov/) to automatically search and extract rare disease epidemiology information from PubMed abstracts.")

#### SIDEBAR WIDGETS ####

#max_results is Maximum number of PubMed ID's to retrieve BEFORE filtering
max_results = st.sidebar.number_input("Maximum number of articles to find in PubMed", min_value=1, max_value=None, value=50)

filtering = st.sidebar.radio("What type of filtering would you like?",('Strict', 'Lenient', 'None')).lower()

extract_diseases = st.sidebar.checkbox("Extract Rare Diseases", value=False)

#### MODEL LOADING ####
@st.experimental_singleton(show_spinner=False)
def load_models_experimental():
    epi_classify = Classify_Pipeline()
    epi_extract = NER_Pipeline()
    rd_identify = GARD_Search()
    return epi_classify, epi_extract, rd_identify

#### DOWNLOAD FUNCTION ####
@st.cache
def convert_df(df):
    # IMPORTANT: Cache the conversion to prevent computation on every rerun
    return df.to_csv().encode('utf-8')

#### SANKEY FUNCTION ####
#@st.cache(allow_output_mutation=True)
@st.experimental_singleton()
def epi_sankey(sankey_data, disease_or_gard_id):
    found, relevant, epidemiologic = sankey_data

    fig = go.Figure(data=[go.Sankey(
        node = dict(
          pad = 15,
          thickness = 20,
          line = dict(color = "white", width = 0.5),
          label = ["PubMed IDs Gathered", "Irrelevant Abstracts","Relevant Abstracts Gathered","Epidemiologic Abstracts","Not Epidemiologic"],
          color = "purple"
        ),
        #label = ["A1", "A2", "B1", "B2", "C1", "C2"]
        link = dict(
          source = [0, 0, 2, 2],
          target = [2, 1, 3, 4],
          value = [relevant, found-relevant, epidemiologic, relevant-epidemiologic]
      ))])
    fig.update_layout(
    hovermode = 'x',
    title="Search for the Epidemiology of "+disease_or_gard_id,
    font=dict(size = 10, color = 'black'),
)

    return fig

#### BEGIN APP ####
with st.spinner('Loading Epidemiology Models and Dependencies...'):
    epi_classify, epi_extract, rd_identify = load_models_experimental()
loaded = st.success('All Models and Dependencies Loaded!')

disease_or_gard_id = st.text_input("Input a rare disease term or NIH GARD ID.")

loaded.empty()

st.markdown("Examples of rare diseases include [**Fellman syndrome**](https://rarediseases.info.nih.gov/diseases/1/gracile-syndrome), [**Classic Homocystinuria**](https://rarediseases.info.nih.gov/diseases/6667/classic-homocystinuria), [**7383**](https://rarediseases.info.nih.gov/diseases/7383/phenylketonuria), and [**GARD:0009941**](https://rarediseases.info.nih.gov/diseases/9941/fshmd1a). A full list of rare diseases tracked by the NIH Genetic and Rare Diseases Information Center (GARD) can be found [here](https://rarediseases.info.nih.gov/diseases/browse-by-first-letter).")

if disease_or_gard_id:
    df, sankey_data, disease_gardid = streamlit_extraction(disease_or_gard_id, max_results, filtering,
                                                        epi_extract, rd_identify, extract_diseases, epi_classify)
    #IF it returns something, then continue.
    if sankey_data:
        df.replace(to_replace='None', value="None")
        st.dataframe(df, height=200)
        csv = convert_df(df)
        #if the user input does not have a number in it (i.e. weak proxy for if it is a GARD ID), then preserve the user input as the disease term.
        disease, gardID = disease_gardid
        if not bool(re.search(r'\d', disease_or_gard_id)):
            disease = disease_or_gard_id
            
        st.download_button(
            label="Download epidemiology results for "+disease+" as CSV",
            data = csv,
            file_name=disease+'.csv',
            mime='text/csv',
            )
        if gardID:
            st.markdown('See the NIH GARD page for ['+disease+'](https://rarediseases.info.nih.gov/diseases/'+str(re.sub('GARD:|0','',gardID))+'/'+str('-'.join(disease.split()))+')')
        
        fig = epi_sankey(sankey_data, disease)
        st.plotly_chart(fig, use_container_width=True)
        
        if 'IDS' in list(df.columns):
            st.markdown('''COLUMNS: \\
                       - PROB_OF_EPI: Probability that the paper is an epidemiologic study based on its abstract. \\
                       - IsEpi: If it is an epidemiologic study (If PROB_OF_EPI >0.5) \\
                       - DIS: Rare disease terms or synonyms identified in the abstract from the GARD Dictionary
                       - IDS: GARD IDs identified in the abstract from the GARD Dictionary \\
                       - EPI: Epidemiology Types are the metrics used to estimate disease burden such as "incidence", "prevalence rate", or "occurrence"
                       - STAT: Epidemiology Rates describe how many people are afflicted by a disease.
                       - DATE: The dates when the epidemiologic studies were conducted
                       - LOC: Where the epidemiologic studies were conducted.
                       - SEX: The biological sexes mentioned in the abstract. Useful for diseases that disproportionately affect one sex over the other or may provide context to composition of the study population
                       - ETHN: Ethnicities, races, and nationalities of those represented in the epidemiologic study.
                    ''')
        else:
            st.subheader("Categories of Results")
            st.markdown("    - **PROB_OF_EPI**: Probability that the paper is an epidemiologic study based on its abstract.  \n  - **IsEpi**: If it is an epidemiologic study (If PROB_OF_EPI >0.5)  \n  - **EPI**: Epidemiology Types are the metrics used to estimate disease burden such as 'incidence', 'prevalence rate', or 'occurrence'  \n  - **STAT**: Epidemiology Rates describe how many people are afflicted by a disease.  \n  - **DATE**: The dates when the epidemiologic studies were conducted  \n  - **LOC**: Where the epidemiologic studies were conducted.  \n  - **SEX**: The biological sexes mentioned in the abstract. Useful for diseases that disproportionately affect one sex over the other or may provide context to composition of the study population  \n  - **ETHN**: Ethnicities, races, and nationalities of those represented in the epidemiologic study.")
        #st.dataframe(data=None, width=None, height=None)