app.py

# -*- coding: utf-8 -*-
"""
Created on Mon Nov 21 16:21:25 2022

@author: luol2
"""

import streamlit as st
from src.nn_model import bioTag_CNN,bioTag_Bioformer
from src.dic_ner import dic_ont
from src.tagging_text import bioTag
import os
import json
from pandas import DataFrame
import nltk 
nltk.download('punkt')
nltk.download('averaged_perceptron_tagger')
nltk.download('wordnet')

st.set_page_config(
    page_title="PhenoTagger",
    page_icon="🎈",
    layout="wide",
    menu_items={
        'Get Help': 'https://www.ncbi.nlm.nih.gov/research/bionlp/',
        'About': "PhenoTagger v1.1"
    }
)


# def _max_width_():
#     max_width_str = f"max-width: 2400px;"
#     st.markdown(
#         f"""
#     <style>
#     .reportview-container .main .block-container{{
#         {max_width_str}
#     }}
#     </style>    
#     """,
#         unsafe_allow_html=True,
#     )


# _max_width_()

# c30, c31, c32 = st.columns([2.5, 1, 3])

# with c30:
#     # st.image("logo.png", width=400)
st.title("👨‍⚕️ PhenoTagger Demo")

with st.expander("ℹ️ - About this app", expanded=True):

    st.write(
        """     
-   This app is an easy-to-use interface built in Streamlit for [PhenoTagger](https://github.com/ncbi-nlp/PhenoTagger) library!
-   PhenoTagger is a hybrid method that combines dictionary and deep learning-based methods to recognize Human Phenotype Ontology (HPO) concepts in unstructured biomedical text. Please refer to [our paper](https://doi.org/10.1093/bioinformatics/btab019) for more details.
-   Contact: [NLM/NCBI BioNLP Research Group](https://www.ncbi.nlm.nih.gov/research/bionlp/)
	    """
    )

    st.markdown("")

st.markdown("")
st.markdown("## 📌 Paste document ")
with st.form(key="my_form"):


    ce, c1, ce, c2, c3 = st.columns([0.07, 1, 0.07, 4, 0.07])
    with c1:
        ModelType = st.radio(
            "Choose your model",
            ["Bioformer(Default)", "CNN"],
            help="Bioformer is more precise, CNN is more efficient",
        )

        if ModelType == "Bioformer(Default)":
            # kw_model = KeyBERT(model=roberta)

            @st.cache(allow_output_mutation=True)
            def load_model():
                ontfiles={'dic_file':'./dict_new/noabb_lemma.dic',
                  'word_hpo_file':'./dict_new/word_id_map.json',
                  'hpo_word_file':'./dict_new/id_word_map.json'}
        

                vocabfiles={'labelfile':'./dict_new/lable.vocab',
                            'config_path':'./vocab/bioformer-cased-v1.0/bert_config.json',
                            'checkpoint_path':'./vocab/bioformer-cased-v1.0/bioformer-cased-v1.0-model.ckpt-2000000',
                            'vocab_path':'./vocab/bioformer-cased-v1.0/vocab.txt'}
                modelfile='./vocab/bioformer_p5n5_b64_1e-5_95_hponew3.h5'
       
    
                biotag_dic=dic_ont(ontfiles)    
    
                nn_model=bioTag_Bioformer(vocabfiles)
                nn_model.load_model(modelfile)
                return nn_model,biotag_dic

            nn_model,biotag_dic = load_model()

        else:
            @st.cache(allow_output_mutation=True)
            def load_model():
                ontfiles={'dic_file':'./dict_new/noabb_lemma.dic',
                  'word_hpo_file':'./dict_new/word_id_map.json',
                  'hpo_word_file':'./dict_new/id_word_map.json'}
        

                vocabfiles={'w2vfile':'./vocab/bio_embedding_intrinsic.d200',   
                            'charfile':'./vocab/char.vocab',
                            'labelfile':'./dict_new/lable.vocab',
                            'posfile':'./vocab/pos.vocab'}
                modelfile='./models/cnn_p5n5_b128_95_hponew1.h5'
        
                biotag_dic=dic_ont(ontfiles)    
            
                nn_model=bioTag_CNN(vocabfiles)
                nn_model.load_model(modelfile)
            
                return nn_model,biotag_dic

            nn_model,biotag_dic = load_model()
        
        para_overlap = st.checkbox(
            "Overlap concept",
            value=True,
            help="Tick this box to identify overlapping concepts",
        )
        para_abbr = st.checkbox(
            "Abbreviaitons",
            value=True,
            help="Tick this box to identify abbreviations",
        )        
        
        para_threshold = st.slider(
            "Threshold",
            min_value=0.5,
            max_value=0.95,
            value=0.95,
            step=0.05,
            help="Retrun the preditions which socre over the threshold.",
        )
      



    with c2:
        doc = st.text_area(
            "Paste your text below",
            height=400,
        )

        # MAX_WORDS = 500
        # import re
        # res = len(re.findall(r"\w+", doc))
        # if res > MAX_WORDS:
        #     st.warning(
        #         "⚠️ Your text contains "
        #         + str(res)
        #         + " words."
        #         + " Only the first 500 words will be reviewed. Stay tuned as increased allowance is coming! 😊"
        #     )

        #     doc = doc[:MAX_WORDS]

        submit_button = st.form_submit_button(label="✨ Submit!")


if not submit_button:
    st.stop()


para_set={
          #model_type':para_model, # cnn or bioformer
          'onlyLongest':para_overlap, # False: return overlap concepts, True only longgest
          'abbrRecog':para_abbr,# False: don't identify abbr, True: identify abbr
          'ML_Threshold':para_threshold,# the Threshold of deep learning model
          }
st.markdown("")
st.markdown("## 💡 Tagging results:")
with st.spinner('Wait for tagging...'):
    tag_result=bioTag(doc,biotag_dic,nn_model,onlyLongest=para_set['onlyLongest'], abbrRecog=para_set['abbrRecog'],Threshold=para_set['ML_Threshold'])

st.markdown('<font style="color: rgb(128, 128, 128);">Move the mouse over the entity to display the HPO id.</font>', unsafe_allow_html=True)
# print('dic...........:',biotag_dic.keys())
# st.write('parameters:', para_overlap,para_abbr,para_threshold)

html_results=''
text_results=doc+'\n'
entity_end=0
hpoid_count={}
if len(tag_result)>=0:
    for ele in tag_result:
        entity_start=int(ele[0])
        html_results+=doc[entity_end:entity_start]
        entity_end=int(ele[1])
        entity_id=ele[2]
        entity_score=ele[3]
        text_results+=ele[0]+'\t'+ele[1]+'\t'+doc[entity_start:entity_end]+'\t'+ele[2]+'\t'+format(float(ele[3]),'.2f')+'\n'
        if entity_id not in hpoid_count.keys():
            hpoid_count[entity_id]=1
        else:
            hpoid_count[entity_id]+=1
        
        html_results+='<font style="background-color: rgb(255, 204, 0)'+';" title="'+entity_id+'">'+doc[entity_start:entity_end]+'</font>'
    html_results+=doc[entity_end:]
                
else:
    html_results=doc
    
st.markdown('<table border="1"><tr><td>'+html_results+'</td></tr></table>', unsafe_allow_html=True)


#table
data_entity=[]
for ele in hpoid_count.keys():
    temp=[ele,biotag_dic.hpo_word[ele][0],hpoid_count[ele]] #hpoid, term name, count
    data_entity.append(temp)


st.markdown("")
st.markdown("")
# st.markdown("## Table output:")

# cs, c1, c2, c3, cLast = st.columns([2, 1.5, 1.5, 1.5, 2])

# with c1:
#     CSVButton2 = download_button(keywords, "Data.csv", "📥 Download (.csv)")
# with c2:
#     CSVButton2 = download_button(keywords, "Data.txt", "📥 Download (.txt)")
# with c3:
#     CSVButton2 = download_button(keywords, "Data.json", "📥 Download (.json)")

# st.header("")

df = (
    DataFrame(data_entity, columns=["HPO_id", "Term name","Frequency"])
    .sort_values(by="Frequency", ascending=False)
    .reset_index(drop=True)
)

df.index += 1

c1, c2, c3 = st.columns([1, 4, 1])

# format_dictionary = {
#     "Relevancy": "{:.1%}",
# }

# df = df.format(format_dictionary)

with c2:
    st.table(df)
    
c1, c2, c3 = st.columns([1, 1, 1])
with c2:
    st.download_button('Download annotations', text_results)