app.py

from Bio.Blast import NCBIWWW, NCBIXML
from Bio.Seq import Seq
from Bio import SeqIO
from Bio.SeqUtils import GC
import time
import warnings
import streamlit as st
from PIL import Image
from io import StringIO

img = Image.open("image.png")

st.image(img, use_column_width=True)

st.write(""" ## **Your results will display here :**
""")
def set_button_style():
    button_style = """
        <style>
            .stButton button {
                background-color: #FF6F61;
                color: white;
                border-radius: 100px;
                border: 2px solid #FFFFFF; 
                box-shadow: 2px 2px 5px #000000;
                width: 160px;
            }
        </style>
    """
    st.markdown(button_style, unsafe_allow_html=True)

    
st.sidebar.title('Protein alignment with BLAST')

up = st.sidebar.file_uploader("Choose an existing fasta file", accept_multiple_files=False)
if up :
    fasta = up.read()


seq = st.sidebar.text_area('Or enter a sequence to analyze : ')

b1 = st.sidebar.button("Calculate GC")
b2 = st.sidebar.button("Replicate")
b3 = st.sidebar.button("Transcribe")
b4 = st.sidebar.button("Back-Transcribe")
b5 = st.sidebar.button("Translate")

type = st.sidebar.text_input('Enter BLAST type : ',placeholder='exemple : blastp, blastn ...')
data = st.sidebar.text_input('Enter the database : ', placeholder='exemple : nt, nr ...')
detailed = st.sidebar.text_input('Enter detailed search : ', placeholder='Homo sapiens')
set_button_style()
b = st.sidebar.button("Run BLAST")
s = st.sidebar.button("Run detailed search")



def blast(seq, type, data):

    st.write("""Starting BLAST search...""") 
    start_time = time.time()
    result_handle2 = NCBIWWW.qblast(type, data, seq)
    end_time = time.time()
    
    st.write("""BLAST search completed.""")
    st.write(f"Time taken: {end_time - start_time:.2f} seconds.")
    
    with open("output.xml", "w") as out_handle:
        out_handle.write(result_handle2.read())
    st.write("Results saved to output.xml")
    st.write("===================================")
    result_handle = open("output.xml")
    blast_record = NCBIXML.read(result_handle)
    st.write(""" **Results :** """)
    count = 0
    for alignment in blast_record.alignments:
        st.write("**"+str(alignment.title)+"**")
        for hsp in alignment.hsps:
            st.write(str(hsp.query[0:50]))
            st.write(str(hsp.query[0:50]))
            st.write(str(hsp.query[0:50]))
            percentage_result = (hsp.query_end - hsp.query_start + 1) / hsp.align_length * 100
            st.write(str(hsp))
            if percentage_result >= 50:  
                st.write(f'<p style="color:green;">{percentage_result:.2f}</p>', unsafe_allow_html=True)
            else:
                st.write(f'<p style="color:red;">{percentage_result:.2f}</p>', unsafe_allow_html=True)
        count += 1
        if count == 10:
            break

def detailed_search(detailed):
    
    result_handle = open("output.xml")
    blast_record = NCBIXML.read(result_handle)
    
    st.write(" Detailed search on : " + "**" + str(detailed) + "**")
    for alignment in blast_record.alignments:
        if detailed in alignment.title:
            st.write(str(alignment.title))
            for hsp in alignment.hsps:
                st.write(str(hsp.query[0:50]))
                st.write(str(hsp.match[0:50]))
                st.write(str(hsp.sbjct[0:50]))
                st.write(str(hsp.positives) + str(hsp.score) + str(hsp.expect) + str(hsp.bits) + str(hsp.gaps))

def save_seq(seq):
    with open("sequenceinput.fsa", "w") as file:
        file.write(str(seq))
    st.write("sequence saved to sequenceinput.fsa")

if seq :
    save_seq(seq)

if (seq and  type  and data and b)  :
    blast(seq, type, data)
elif (up and type and data and b) :
    blast(fasta, type, data)
 
if detailed and s : 
    detailed_search(detailed)


def calculate(seq):
    with warnings.catch_warnings():
        warnings.simplefilter("ignore")  # Ignore all warnings
        gc_content = GC(seq)
    st.write(f"### **GC :** {gc_content}")

if seq and b1 :
    calculate(seq)
elif up and b1 :
    calculate(fasta.decode())

def reverse(seq):
    sequence = Seq(seq)
    r_seq = sequence.reverse_complement()
    chunked_r_seq = [r_seq[i:i+50] for i in range(0, len(r_seq), 50)]

    st.write("### **Reversed sequence :**")
    for chunk in chunked_r_seq:
        st.write("#### " + str(chunk))

if seq and b2 :
    reverse(seq)
elif up and b2 : 
    reverse(fasta)

def transcribe_seq(seq):
    sequence = Seq(seq)
    r_seq = sequence.transcribe()
    chunked_r_seq = [r_seq[i:i+50] for i in range(0, len(r_seq), 50)]

    st.write("### **Transcribed sequence :** ")
    for chunk in chunked_r_seq:
        st.write("#### " + str(chunk))

if seq and b3 :
    transcribe_seq(seq)
elif up and b3 :
    transcribe_seq(fasta)

def back_transcribe_seq(seq):
    sequence = Seq(seq)
    r_seq = sequence.back_transcribe()
    chunked_r_seq = [r_seq[i:i+50] for i in range(0, len(r_seq), 50)]

    st.write("### **Back-Transcribed sequence :** ")
    for chunk in chunked_r_seq:
        st.write("#### " + str(chunk))

if seq and b4 :
    back_transcribe_seq(seq)
elif up and b4 :
    back_transcribe_seq(fasta)

def translate_seq(seq):
    try :
        sequence = Seq(seq)
        r_seq = sequence.translate()
        chunked_r_seq = [r_seq[i:i+50] for i in range(0, len(r_seq), 50)]

        st.write("### **Translated sequence :** ")    
        for chunk in chunked_r_seq:
            st.write("#### " + str(chunk))
    except Exception as e:
        st.write(f"#### Unable to translate this sequence! {e}")

if seq and b5 :
    translate_seq(seq)
elif up and b5 : 
    transcribe_seq(fasta)