Spaces:
Sleeping
Sleeping
File size: 6,634 Bytes
444a81b 33d056e 444a81b 33d056e 444a81b 33d056e 444a81b |
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 |
import os
os.system("pip install streamlit pandas xlsxwriter openpyxl")
import streamlit as st
import pandas as pd
import xlsxwriter
from io import BytesIO
from collections import defaultdict
def find_homorepeats(protein):
n = len(protein)
freq = defaultdict(int)
i = 0
while i < n:
curr = protein[i]
repeat = ""
while i < n and curr == protein[i]:
repeat += protein[i]
i += 1
if len(repeat) > 1:
freq[repeat] += 1
return freq
def find_hetero_amino_acid_repeats(sequence):
repeat_counts = defaultdict(int)
for length in range(2, len(sequence) + 1):
for i in range(len(sequence) - length + 1):
substring = sequence[i:i+length]
repeat_counts[substring] += 1
return {k: v for k, v in repeat_counts.items() if v > 1}
def fragment_protein_sequence(sequence, max_length=1000):
return [sequence[i:i+max_length] for i in range(0, len(sequence), max_length)]
def check_boundary_repeats(fragments, final_repeats, overlap=50):
for i in range(len(fragments) - 1):
left_overlap = fragments[i][-overlap:]
right_overlap = fragments[i + 1][:overlap]
overlap_region = left_overlap + right_overlap
boundary_repeats = find_hetero_amino_acid_repeats(overlap_region)
for substring, count in boundary_repeats.items():
if any(aa in left_overlap for aa in substring) and any(aa in right_overlap for aa in substring):
final_repeats[substring] += count
return final_repeats
def find_new_boundary_repeats(fragments, final_repeats, overlap=50):
new_repeats = defaultdict(int)
for i in range(len(fragments) - 1):
left_overlap = fragments[i][-overlap:]
right_overlap = fragments[i + 1][:overlap]
overlap_region = left_overlap + right_overlap
boundary_repeats = find_hetero_amino_acid_repeats(overlap_region)
for substring, count in boundary_repeats.items():
if any(aa in left_overlap for aa in substring) and any(aa in right_overlap for aa in substring):
if substring not in final_repeats:
new_repeats[substring] += count
return new_repeats
def process_protein_sequence(sequence, analysis_type, overlap=50):
fragments = fragment_protein_sequence(sequence)
final_repeats = defaultdict(int)
homo_repeats = {}
for fragment in fragments:
if analysis_type in ["Hetero", "Both"]:
fragment_repeats = find_hetero_amino_acid_repeats(fragment)
for k, v in fragment_repeats.items():
final_repeats[k] += v
if analysis_type in ["Hetero", "Both"]:
final_repeats = check_boundary_repeats(fragments, final_repeats, overlap)
new_repeats = find_new_boundary_repeats(fragments, final_repeats, overlap)
for k, v in new_repeats.items():
final_repeats[k] += v
if analysis_type in ["Homo", "Both"]:
homo_repeats = find_homorepeats(sequence)
for k, v in homo_repeats.items():
final_repeats[k] += v
if analysis_type == "Hetero" or "Both":
for k, v in homo_repeats.items():
if k in final_repeats:
final_repeats[k] -= v
if final_repeats[k] <= 0:
del final_repeats[k]
return final_repeats
def process_excel(excel_data, analysis_type):
repeats = set()
sequence_data = []
for sheet_name in excel_data.sheet_names:
df = excel_data.parse(sheet_name)
if len(df.columns) < 3:
st.error(f"Error: The sheet '{sheet_name}' must have at least three columns: ID, Protein Name, Sequence")
return None, None
for _, row in df.iterrows():
entry_id = str(row[0])
protein_name = str(row[1])
sequence = str(row[2]).replace('"', '').replace(' ', '')
freq = process_protein_sequence(sequence, analysis_type)
sequence_data.append((entry_id, protein_name, freq))
repeats.update(freq.keys())
return repeats, sequence_data
def create_excel(sequences_data, repeats, filenames):
output = BytesIO()
workbook = xlsxwriter.Workbook(output, {'in_memory': True})
for file_index, file_data in enumerate(sequences_data):
filename = filenames[file_index]
worksheet = workbook.add_worksheet(filename[:31])
worksheet.write(0, 0, "Entry ID")
worksheet.write(0, 1, "Protein Name")
col = 2
for repeat in sorted(repeats):
worksheet.write(0, col, repeat)
col += 1
row = 1
for entry_id, protein_name, freq in file_data:
worksheet.write(row, 0, entry_id)
worksheet.write(row, 1, protein_name)
col = 2
for repeat in sorted(repeats):
worksheet.write(row, col, freq.get(repeat, 0))
col += 1
row += 1
workbook.close()
output.seek(0)
return output
st.title("Protein Repeat Analysis")
analysis_type = st.radio("Select analysis type:", ["Homo", "Hetero", "Both"], index=2)
uploaded_files = st.file_uploader("Upload Excel files", accept_multiple_files=True, type=["xlsx"])
if uploaded_files:
all_repeats = set()
all_sequences_data = []
filenames = []
for file in uploaded_files:
excel_data = pd.ExcelFile(file)
repeats, sequence_data = process_excel(excel_data, analysis_type)
if repeats is not None:
all_repeats.update(repeats)
all_sequences_data.append(sequence_data)
filenames.append(file.name)
if all_sequences_data:
st.success(f"Processed {len(uploaded_files)} files successfully!")
excel_file = create_excel(all_sequences_data, all_repeats, filenames)
st.download_button(
label="Download Excel file",
data=excel_file,
file_name="protein_repeat_results.xlsx",
mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
)
if st.checkbox("Show Results Table"):
rows = []
for file_index, file_data in enumerate(all_sequences_data):
filename = filenames[file_index]
for entry_id, protein_name, freq in file_data:
row = {"Filename": filename, "Entry ID": entry_id, "Protein Name": protein_name}
row.update({repeat: freq.get(repeat, 0) for repeat in sorted(all_repeats)})
rows.append(row)
result_df = pd.DataFrame(rows)
st.dataframe(result_df)
|