app.py

# Import libraries
import streamlit as st
import pandas as pd
from Bio import SeqIO
from Bio.SeqUtils.ProtParam import ProteinAnalysis
from Bio.Graphics.GenomeDiagram import Diagram, Track, FeatureSet
from reportlab.lib import colors
from reportlab.lib.units import cm
from io import StringIO
from collections import Counter
import numpy as np
import altair as alt
import os
import re
import plotly.express as px

# Ensure the 'temp' directory exists for saving temporary files
temp_dir = "temp"
os.makedirs(temp_dir, exist_ok=True)

# Function to parse GenBank file
def parse_genbank(uploaded_file):
    stringio = StringIO(uploaded_file.getvalue().decode("utf-8"))
    record = SeqIO.read(stringio, "genbank")
    organism = record.annotations.get('organism', 'Unknown Organism')
    features = record.features
    feature_types = Counter([feature.type for feature in features])

    genes, cds = [], []
    for feature in features:
        if feature.type == "gene":
            genes.append(feature)
        elif feature.type == "CDS":
            cds.append(feature)

    gene_info = [{
        'Gene': gene_feature.qualifiers.get('gene', ['N/A'])[0],
        'Length': len(gene_feature),
        'Location': str(gene_feature.location),
        'Sequence': str(gene_feature.extract(record.seq))
    } for gene_feature in genes]

    cds_info = [{
        'Gene': cds_feature.qualifiers.get('gene', ['N/A'])[0],
        'Protein': cds_feature.qualifiers.get('translation', ['N/A'])[0],
        'Length': len(cds_feature),
        'Location': str(cds_feature.location)
    } for cds_feature in cds]

    gc_content = (str(record.seq).count('G') + str(record.seq).count('C')) / len(record.seq) * 100

    return organism, gene_info, cds_info, gc_content, len(record.seq), feature_types, str(record.seq)

# Function to calculate GC content over genome
def calculate_gc_content(sequence, window_size=1000):
    gc_content = [
        (sequence[i:i+window_size].count('G') + sequence[i:i+window_size].count('C')) / window_size * 100
        for i in range(0, len(sequence) - window_size + 1, window_size)
    ]
    return gc_content

# Function to calculate k-mers
def calculate_kmers(sequence, k):
    kmers = Counter([sequence[i:i+k] for i in range(len(sequence) - k + 1)])
    return kmers

# Function to add molecular weight and isoelectric point to CDS information
def add_protein_features(cds_info):
    for cds in cds_info:
        if cds['Protein'] != 'N/A':
            prot_analysis = ProteinAnalysis(cds['Protein'])
            cds['Molecular Weight'] = prot_analysis.molecular_weight()
            cds['Isoelectric Point'] = prot_analysis.isoelectric_point()
        else:
            cds['Molecular Weight'] = 'N/A'
            cds['Isoelectric Point'] = 'N/A'
    return cds_info

# Updated Function to generate genome diagram
def create_genome_diagram(genbank_content, output_file_path, colors_dict, diagram_type="linear", diagram_size=(30, 10)):
    from Bio.Graphics.GenomeDiagram import Diagram, Track, FeatureSet
    record = SeqIO.read(StringIO(genbank_content), "genbank")
    gd_diagram = Diagram(record.id)
    
    # Create separate tracks for different feature types
    max_tracks = len(colors_dict)
    track_indices = {feature_type: idx+1 for idx, feature_type in enumerate(colors_dict.keys())}
    feature_tracks = {}
    feature_sets = {}  # Dictionary to store FeatureSets
    for feature_type, idx in track_indices.items():
        feature_tracks[feature_type] = gd_diagram.new_track(
            idx, name=feature_type, scale=False, greytrack=False, height=0.5
        )
        # Store the FeatureSet
        feature_sets[feature_type] = feature_tracks[feature_type].new_set()
    
    for feature in record.features:
        feature_type = feature.type
        if feature_type in colors_dict:
            color = colors.HexColor(colors_dict[feature_type])
            # Retrieve the FeatureSet from the dictionary
            feature_set = feature_sets[feature_type]
            feature_set.add_feature(
                feature,
                color=color,
                label=True,
                label_size=6,  # Decreased label size
                label_angle=0,
                label_position="start",  # Position label at the start of the feature
                label_strand=1 if feature.strand == 1 else -1,  # Position labels according to strand
                sigil="ARROW",  # Use arrows to represent features
                arrowshaft_height=1.0,
                arrowhead_length=1.0,
            )
    
    if diagram_type.lower() == "circular":
        gd_diagram.draw(
            format="circular",
            circular=True,
            pagesize=(diagram_size[0]*cm, diagram_size[1]*cm),
            start=0,
            end=len(record),
            circle_core=0.7
        )
    else:
        gd_diagram.draw(
            format="linear",
            pagesize=(diagram_size[0]*cm, diagram_size[1]*cm),
            fragments=1,
            start=0,
            end=len(record),
            tracklines=False  # Remove track lines to reduce clutter
        )
    gd_diagram.write(output_file_path, "SVG")

# Function to search for a motif or pattern within the DNA sequence
def search_motif(sequence, motif):
    matches = [(match.start(), match.end()) for match in re.finditer(motif, sequence)]
    return matches

# Function to calculate codon usage frequencies
def calculate_codon_usage(sequence):
    codons = [sequence[i:i+3] for i in range(0, len(sequence)-2, 3)]
    codons = [codon for codon in codons if len(codon) == 3]  # Ensure codons are of length 3
    codon_freq = Counter(codons)
    total_codons = sum(codon_freq.values())
    codon_freq_percent = {codon: (count / total_codons) * 100 for codon, count in codon_freq.items()}
    return codon_freq_percent

# Streamlit UI setup
st.set_page_config(page_title="Genomic Data Dashboard", page_icon="🧬", layout="wide")
uploaded_file = st.file_uploader("Upload a GenBank file", type=['gb', 'gbk'])

if uploaded_file is not None:
    organism, gene_info, cds_info, gc_content, sequence_length, feature_types, sequence = parse_genbank(uploaded_file)
    cds_info = add_protein_features(cds_info)
    gene_df = pd.DataFrame(gene_info)
    cds_df = pd.DataFrame(cds_info)

    # Sidebar
    with st.sidebar:
        st.title('Genomic Data Dashboard')
        st.write(f'**Organism:** {organism}')
        window_size = st.number_input('GC content sliding window size', min_value=100, max_value=10000, value=1000)
        k = st.number_input('k-mer size', min_value=1, max_value=10, value=6)
        motif = st.text_input("Enter motif or pattern to search")

        st.markdown("### Genome Diagram Settings")
        diagram_type = st.radio("Select Diagram Type:", ["Linear", "Circular"], index=0)
        diagram_width = st.number_input("Diagram Width (cm):", min_value=1, max_value=50, value=30)
        diagram_height = st.number_input("Diagram Height (cm):", min_value=1, max_value=50, value=10)

        # Sidebar options for diagram customization
        st.markdown("### Feature Colors")
        color_gene = st.color_picker("Pick a color for genes", '#ff9999')
        color_cds = st.color_picker("Pick a color for CDS", '#66b3ff')
        color_trna = st.color_picker("Pick a color for tRNA", '#99ff99')
        color_rrna = st.color_picker("Pick a color for rRNA", '#ffcc99')

    # Main content
    col1, col2 = st.columns(2)
    with col1:
        st.markdown('### General Information')
        st.write(f'**Organism:** {organism}')
        st.write(f'**Sequence Length:** {sequence_length} bp')
        st.write(f'**GC Content:** {gc_content:.2f}%')
        st.write(f'**Number of Genes:** {len(gene_df)}')
        st.write(f'**Number of Coding Sequences (CDS):** {len(cds_df)}')
        st.markdown('### Feature Counts')
        for feature_type, count in feature_types.items():
            st.write(f"**{feature_type}:** {count}")

    with col2:
        st.markdown('### GC Content Over Genome')
        gc_content_over_genome = calculate_gc_content(sequence, window_size)
        gc_chart = alt.Chart(pd.DataFrame({
            'GC Content': gc_content_over_genome,
            'Position': np.arange(len(gc_content_over_genome)) * window_size
        })).mark_line().encode(
            x='Position:Q',
            y='GC Content:Q'
        ).properties(height=200)
        st.altair_chart(gc_chart, use_container_width=True)

        st.markdown('### K-mer Analysis')
        kmers = calculate_kmers(sequence, k)
        kmer_df = pd.DataFrame.from_dict(kmers, orient='index', columns=['Frequency']).sort_values('Frequency', ascending=False).head(20)
        st.bar_chart(kmer_df)

    # Construct the colors dictionary
    feature_colors = {
        'gene': color_gene,
        'CDS': color_cds,
        'tRNA': color_trna,
        'rRNA': color_rrna
    }

    # Generate and display genome diagram with user-selected feature colors
    output_file_path_svg = os.path.join(temp_dir, "genome_diagram.svg")
    create_genome_diagram(
        uploaded_file.getvalue().decode("utf-8"),
        output_file_path_svg,
        feature_colors,
        diagram_type=diagram_type.lower(),
        diagram_size=(diagram_width, diagram_height)
    )
    st.image(output_file_path_svg, caption='Genome Diagram')

    # Motif Search
    if motif:
        matches = search_motif(sequence, motif)
        st.markdown(f"### Motif Search Results for '{motif}'")
        if matches:
            st.write("Matches found at positions:")
            matches_df = pd.DataFrame(matches, columns=["Start", "End"])
            st.dataframe(matches_df)
        else:
            st.write("No matches found.")

    # Codon Usage Analysis
    st.markdown("### Codon Usage Analysis")
    codon_usage_freq = calculate_codon_usage(sequence)
    codon_usage_df = pd.DataFrame.from_dict(codon_usage_freq, orient='index', columns=['Frequency (%)'])
    codon_usage_df.index.name = 'Codon'
    codon_usage_df.reset_index(inplace=True)
    st.dataframe(codon_usage_df)

    # Interactive Visualization
    st.markdown("### Interactive Visualization")
    fig = px.bar(
        codon_usage_df,
        x='Codon',
        y='Frequency (%)',
        labels={'Codon': 'Codon', 'Frequency (%)': 'Frequency (%)'},
        title="Codon Usage Frequency"
    )
    st.plotly_chart(fig)

    # Additional Information
    with st.expander("View All Genes"):
        st.dataframe(gene_df)
    with st.expander("View All Coding Sequences"):
        st.dataframe(cds_df[['Gene', 'Length', 'Molecular Weight', 'Isoelectric Point']])
else:
    st.warning("Please upload a GenBank file.")

# Add copyright information section at the end of the main page
st.markdown("""
---
**Copyright Notice**: © 2024 Dr. Yash Munnalal Gupta. All rights reserved.

For inquiries or permissions, contact: [yash.610@live.com](mailto:yash.610@live.com)
""", unsafe_allow_html=True)