app.py

import streamlit as st
import numpy as np
import pandas as pd

# Mutation site headers removed 3614,
mutation_site_headers_actual = [
    3244, 3297, 3350, 3399, 3455, 3509, 3562, 
    3665, 3720, 3773, 3824, 3879, 3933, 3985, 4039,
    4089, 4145, 4190, 4245, 4298, 4349, 4402, 4455,
    4510, 4561, 4615, 4668, 4720, 4773, 4828, 4882
]
# Thresholds for each mutation site removed 3614: 0.091557752,
thresholds_actual = pd.Series({
    3244: 1.096910677, 3297: 0.923658795, 3350: 0.668939037, 3399: 0.914305214,
    3455: 1.297392984, 3509: 1.812636208, 3562: 1.185047484, 
    3665: 0.298007308, 3720: 0.58857544, 3773: 0.882561082, 3824: 1.149082617,
    3879: 0.816050702, 3933: 2.936517653, 3985: 1.597166791, 4039: 0.962108082,
    4089: 1.479783497, 4145: 0.305853225, 4190: 1.311869541, 4245: 1.707556905,
    4298: 0.875013076, 4349: 1.227704526, 4402: 0.593206446, 4455: 1.179633137,
    4510: 1.272477799, 4561: 1.293841573, 4615: 1.16821885, 4668: 1.40306,
    4720: 0.706530878, 4773: 1.483114072, 4828: 0.954939873, 4882: 1.47524328
})

# Mutation site headers reordered: 4402 to 3244, 4882 to 4455
mutation_site_headers = [
    4402, 4349, 4298, 4245, 4190, 4145, 4089, 4039,
    3985, 3933, 3879, 3824, 3773, 3720, 3665,
    3562, 3509, 3455, 3399, 3350, 3297, 3244,  # 1–23
    4882, 4828, 4773, 4720, 4668, 4615, 4561, 4510, 4455   # 24–32
]

# Thresholds reordered accordingly
thresholds = pd.Series({h: thresholds_actual[h] for h in mutation_site_headers})

# === Utility functions ===

# Voyager ASCII 6-bit conversion table
voyager_table = {
    i: ch for i, ch in enumerate([
        ' ', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I',
        'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S',
        'T', 'U', 'V', 'W', 'X', 'Y', 'Z', '0', '1', '2',
        '3', '4', '5', '6', '7', '8', '9', '.', ',', '(', 
        ')','+', '-', '*', '/', '=', '$', '!', ':', '%', 
        '"', '#', '@', "'", '?', '&'
    ])
}
reverse_voyager_table = {v: k for k, v in voyager_table.items()}

def string_to_binary_labels(s: str) -> list[int]:
    bits = []
    for char in s:
        val = reverse_voyager_table.get(char.upper(), 0)
        char_bits = [(val >> bit) & 1 for bit in range(5, -1, -1)]
        bits.extend(char_bits)
    return bits

def binary_labels_to_string(bits: list[int]) -> str:
    chars = []
    for i in range(0, len(bits), 6):
        chunk = bits[i:i+6]
        if len(chunk) < 6:
            chunk += [0] * (6 - len(chunk))
        val = sum(b << (5 - j) for j, b in enumerate(chunk))
        chars.append(voyager_table.get(val, '?'))
    return ''.join(chars)


# === Streamlit App ===

st.title("ASCII & Binary Label Converter")
tab1, tab2, tab3, tab4, tab5 = st.tabs(["Text to Binary Labels (31)", "EF → Binary → String (31)", "Text to Binary Labels (32)", "EF → Binary (32)", "Binary → String"])

# Tab 1: Text to Binary
with tab1:
    user_input = st.text_input("Enter text", value="DNA", key="input_text_31")
    if user_input:
        ascii_codes = [reverse_voyager_table.get(c.upper(), 0) for c in user_input]
        binary_labels = string_to_binary_labels(user_input)

        # st.subheader("Voyager ASCII Codes")
        # st.write(ascii_codes)

        st.subheader("Binary Labels per Character")
        grouped = [binary_labels[i:i+6] for i in range(0, len(binary_labels), 6)]
        for i, bits in enumerate(grouped):
            st.write(f"'{user_input[i]}' → {bits}")

        st.subheader("Binary Labels (31-bit groups)")
        groups = []
        for i in range(0, len(binary_labels), 31):
            group = binary_labels[i:i+31]
            group += [0] * (31 - len(group))
            groups.append(group + [sum(group)])

        df = pd.DataFrame(groups, columns=[str(h) for h in mutation_site_headers] + ["Edited Sites"])
        st.dataframe(df)
        st.download_button("Download as CSV", df.to_csv(index=False), "text_31_binary_labels.csv", key="download_csv_tab1_31csv")

        ascending_headers = sorted(mutation_site_headers_actual)
        df_sorted = df[[str(h) for h in ascending_headers if str(h) in df.columns]].copy()
        
        if "3614" not in df_sorted.columns:
            idx = df_sorted.columns.get_loc("3562") + 1  # Insert after 3562
            df_sorted.insert(idx, "3614", 0)

        st.subheader("Binary Labels (Ascending Order 3244 → 4882)")
        st.dataframe(df_sorted)
        st.download_button("Download Ascending Order CSV", df_sorted.to_csv(index=False), "text_binary_labels_ascending.csv", key="download_csv_tab1_ascend")

        # === Robot Preparation Script Generation ===
        st.subheader("Robot Preparation Script")
        robot_template = pd.read_csv("/home/user/app/Robot.csv", skiprows=3)
        robot_template.columns = ['Labware', 'Source', 'Labware_2', 'Destination', 'Volume', 'Tool', 'Name']

        # Add Sample numbers for well referencing
        df_sorted.insert(0, 'Sample', range(1, len(df_sorted)+1))

        # Step 1: Count the number of edited sites per row
        df_sorted['# donors'] = df_sorted.iloc[:, 1:].sum(axis=1)

        # Step 2: Calculate volume per donor (32 / # donors)
        df_sorted['volume donors (µl)'] = 32 / df_sorted['# donors']

        # Step 3: Generate the robot script
        robot_script = []
        source_wells = robot_template['Source'].unique().tolist()
        if len(source_wells) < 32:
            source_wells += [f"Fake{i}" for i in range(32 - len(source_wells))]
        source_wells = source_wells[:32]


        st.write(f"Number of source wells: {len(source_wells)}")
        st.write(f"Number of binary columns: {len(df_sorted.columns[1:33])}")

        for i, col in enumerate(df_sorted.columns[1:33]):
            for row_idx, sample in df_sorted.iterrows():
                if sample[col] == 1:
                    source = source_wells[i]
                    dest = f"A{sample['Sample']}"
                    vol = round(sample['volume donors (µl)'], 2)
                    robot_script.append({'Source': source, 'Destination': dest, 'Volume': vol})

        robot_script_df = pd.DataFrame(robot_script)
        st.dataframe(robot_script_df)
        st.download_button("Download Robot Script CSV", robot_script_df.to_csv(index=False), "robot_script.csv", key="download_csv_tab1_robot")

        # === Robot Preparation Script (Based on Ascending Order 3244 → 4882) ===
        st.subheader("Robot Preparation Script (Based on Ascending Order 3244 → 4882)")
        
        df_sorted_ascend =df[[str(h) for h in ascending_headers if str(h) in df.columns]].copy()
        
        # Only insert 'Sample' if it doesn't already exist
        if 'Sample' not in df_sorted_ascend.columns:
            df_sorted_ascend.insert(0, 'Sample', range(1, len(df_sorted_ascend)+1))
        
        # Recalculate donors and volume
        df_sorted_ascend['# donors'] = df_sorted_ascend.iloc[:, 1:].sum(axis=1)
        df_sorted_ascend['volume donors (µl)'] = 32 / df_sorted_ascend['# donors']

        st.subheader("df_sorted_ascend")
        st.write(df_sorted_ascend)
        
        robot_script_ascend = []
        # Use the same 32 source wells
        for i, col in enumerate(df_sorted_ascend.columns[1:33]):
            for row_idx, sample in df_sorted_ascend.iterrows():
                if sample[col] == 1:
                    source = source_wells[i]
                    dest = f"A{sample['Sample']}"
                    vol = round(sample['volume donors (µl)'], 2)
                    robot_script_ascend.append({'Source': source, 'Destination': dest, 'Volume': vol})
        
        robot_script_ascend_df = pd.DataFrame(robot_script_ascend)
        st.dataframe(robot_script_ascend_df)
        st.download_button("Download Ascending Robot Script CSV", robot_script_ascend_df.to_csv(index=False), "robot_script_ascending.csv", key="download_csv_tab1_robot_ascend")

# Tab 2: EF → Binary
with tab2:
    st.write("Upload an Editing Frequency CSV or enter manually:")
    st.write("**Note:** Please upload CSV files **without column headers**, in ascending order from 3244 to 4882.")
    ef_file = st.file_uploader("Upload EF CSV", type=["csv"], key="ef")

    if ef_file:
        ef_df = pd.read_csv(ef_file, header=None)
        ef_df.columns = [str(site) for site in sorted(mutation_site_headers_actual)]
    else:
        ef_df = pd.DataFrame(columns=[str(site) for site in sorted(mutation_site_headers_actual)])

    edited_df = st.data_editor(ef_df, num_rows="dynamic")

    if st.button("Convert to Binary Labels", key="convert_button_tab2"):
        binary_part = pd.DataFrame()
        for col in sorted(mutation_site_headers_actual):
            col_str = str(col)
            threshold = thresholds_actual[col]
            binary_part[col_str] = (edited_df[col_str].astype(float) >= threshold).astype(int)

        binary_reordered = binary_part[[str(h) for h in mutation_site_headers if str(h) in binary_part.columns]]

        def color_binary(val):
            if val == 1: return "background-color: lightgreen"
            if val == 0: return "background-color: lightcoral"
            return ""

        st.subheader("Binary Labels (Reordered 4402→3244, 4882→4455)")
        styled = binary_reordered.style.applymap(color_binary)
        st.dataframe(styled)
        st.download_button("Download CSV", binary_reordered.to_csv(index=False), "ef_binary_labels.csv", key="download_csv_tab2_csv")

        all_bits = binary_reordered.values.flatten().tolist()
        decoded_string = binary_labels_to_string(all_bits)
        st.subheader("Decoded String (continuous across rows)")
        st.write(decoded_string)

        st.subheader("Binary Labels (Ascending 3244→4882)")
        st.dataframe(binary_part.style.applymap(color_binary))
        st.download_button("Download Ascending Order CSV", binary_part.to_csv(index=False), "ef_binary_labels_ascending.csv", key="download_csv_tab2_ascend")

        all_bits = binary_part.values.flatten().tolist()
        decoded_string = binary_labels_to_string(all_bits)
        st.subheader("Decoded String (continuous across rows)")
        st.write(decoded_string)


# Mutation site headers did not remove 3614,
mutation_site_headers_actual_3614 = [
    3244, 3297, 3350, 3399, 3455, 3509, 3562, 3614,
    3665, 3720, 3773, 3824, 3879, 3933, 3985, 4039,
    4089, 4145, 4190, 4245, 4298, 4349, 4402, 4455,
    4510, 4561, 4615, 4668, 4720, 4773, 4828, 4882
]
# Thresholds for each mutation site removed 3614: 0.091557752,
thresholds_actual_3614 = pd.Series({
    3244: 1.096910677, 3297: 0.923658795, 3350: 0.668939037, 3399: 0.914305214,
    3455: 1.297392984, 3509: 1.812636208, 3562: 1.185047484, 3614: 0.157969131375,
    3665: 0.298007308, 3720: 0.58857544, 3773: 0.882561082, 3824: 1.149082617,
    3879: 0.816050702, 3933: 2.936517653, 3985: 1.597166791, 4039: 0.962108082,
    4089: 1.479783497, 4145: 0.305853225, 4190: 1.311869541, 4245: 1.707556905,
    4298: 0.875013076, 4349: 1.227704526, 4402: 0.593206446, 4455: 1.179633137,
    4510: 1.272477799, 4561: 1.293841573, 4615: 1.16821885, 4668: 1.40306,
    4720: 0.706530878, 4773: 1.483114072, 4828: 0.954939873, 4882: 1.47524328
})

# Mutation site headers reordered: 4402 to 3244, 4882 to 4455
mutation_site_headers_3614 = [
    4402, 4349, 4298, 4245, 4190, 4145, 4089, 4039,
    3985, 3933, 3879, 3824, 3773, 3720, 3665, 3614,
    3562, 3509, 3455, 3399, 3350, 3297, 3244,  # 1–23
    4882, 4828, 4773, 4720, 4668, 4615, 4561, 4510, 4455   # 24–32
]

# Thresholds reordered accordingly
thresholds_3614 = pd.Series({h: thresholds_actual_3614[h] for h in mutation_site_headers_3614})

# === Utility functions ===

# Voyager ASCII 6-bit conversion table
voyager_table = {
    i: ch for i, ch in enumerate([
        ' ', 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'I',
        'J', 'K', 'L', 'M', 'N', 'O', 'P', 'Q', 'R', 'S',
        'T', 'U', 'V', 'W', 'X', 'Y', 'Z', '0', '1', '2',
        '3', '4', '5', '6', '7', '8', '9', '.', '(', ')',
        '+', '-', '*', '/', '=', '$', '!', ':', '%', '"',
        '#', '@', '\'', '?', '&'
    ])
}
reverse_voyager_table = {v: k for k, v in voyager_table.items()}


# Tab 3: Text to Binary (32)
with tab3:
    user_input_32 = st.text_input("Enter text", value="DNA", key="input_text_32")
    if user_input_32:
        ascii_codes = [ord(c) for c in user_input_32]
        binary_labels = string_to_binary_labels(user_input_32)

        st.subheader("ASCII Codes")
        st.write(ascii_codes)

        st.subheader("Binary Labels per Character")
        grouped = [binary_labels[i:i+6] for i in range(0, len(binary_labels), 6)]
        for i, bits in enumerate(grouped):
            st.write(f"'{user_input_32[i]}' → {bits}")

        st.subheader("Binary Labels (32-bit groups)")
        groups = []
        for i in range(0, len(binary_labels), 32):
            group = binary_labels[i:i+32]
            group += [0] * (32 - len(group))
            groups.append(group + [sum(group)])

        df = pd.DataFrame(groups, columns=[str(h) for h in mutation_site_headers_3614] + ["Edited Sites"])
        st.dataframe(df)
        st.download_button("Download as CSV", df.to_csv(index=False), "text_32_binary_labels.csv", key="download_csv_tab3_csv")

        ascending_headers = sorted(mutation_site_headers_actual_3614)
        df_sorted = df[[str(h) for h in ascending_headers if str(h) in df.columns]]
        st.subheader("Binary Labels (Ascending Order 3244 → 4882)")
        st.dataframe(df_sorted)
        st.download_button("Download Ascending Order CSV", df_sorted.to_csv(index=False), "text_binary_labels_ascending.csv", key="download_csv_tab3_ascend")

        # === Robot Preparation Script Generation ===
        st.subheader("Robot Preparation Script")
        robot_template = pd.read_csv("/home/user/app/Robot.csv", skiprows=3)
        robot_template.columns = ['Labware', 'Source', 'Labware_2', 'Destination', 'Volume', 'Tool', 'Name']

        # Add Sample numbers for well referencing
        df_sorted.insert(0, 'Sample', range(1, len(df_sorted)+1))

        # Step 1: Count the number of edited sites per row
        df_sorted['# donors'] = df_sorted.iloc[:, 1:].sum(axis=1)

        # Step 2: Calculate volume per donor (32 / # donors)
        df_sorted['volume donors (µl)'] = 32 / df_sorted['# donors']

        # Step 3: Generate the robot script
        robot_script = []
        source_wells = robot_template['Source'].unique().tolist()[:32]

        for i, col in enumerate(df_sorted.columns[1:33]):
            for row_idx, sample in df_sorted.iterrows():
                if sample[col] == 1:
                    source = source_wells[i]
                    dest = f"A{sample['Sample']}"
                    vol = round(sample['volume donors (µl)'], 2)
                    robot_script.append({'Source': source, 'Destination': dest, 'Volume': vol})

        robot_script_df = pd.DataFrame(robot_script)
        st.dataframe(robot_script_df)
        st.download_button("Download Robot Script CSV", robot_script_df.to_csv(index=False), "robot_script.csv", key="download_csv_tab3_robot")


# Tab 4: EF → Binary (32)
with tab4:
    st.write("Upload an Editing Frequency CSV or enter manually:")
    st.write("**Note:** Please upload CSV files **without column headers**, in ascending order from 3244 to 4882.")
    ef_file_2 = st.file_uploader("Upload EF CSV", type=["csv"], key="ef2")

    if ef_file_2:
        ef_df = pd.read_csv(ef_file_2, header=None)
        ef_df.columns = [str(site) for site in sorted(mutation_site_headers_actual_3614)]
    else:
        ef_df = pd.DataFrame(columns=[str(site) for site in sorted(mutation_site_headers_actual_3614)])

    edited_df = st.data_editor(ef_df, num_rows="dynamic")

    if st.button("Convert to Binary Labels", key="convert_button_tab4"):
        binary_part = pd.DataFrame()
        for col in sorted(mutation_site_headers_actual_3614):
            col_str = str(col)
            threshold = thresholds_actual_3614[col]
            binary_part[col_str] = (edited_df[col_str].astype(float) >= threshold).astype(int)

        binary_reordered = binary_part[[str(h) for h in mutation_site_headers_3614 if str(h) in binary_part.columns]]

        def color_binary(val):
            if val == 1: return "background-color: lightgreen"
            if val == 0: return "background-color: lightcoral"
            return ""

        st.subheader("Binary Labels (Reordered 4402→3244, 4882→4455)")
        styled = binary_reordered.style.applymap(color_binary)
        st.dataframe(styled)
        st.download_button("Download CSV", binary_reordered.to_csv(index=False), "ef_binary_labels.csv", key="download_csv_tab4_csv")

        all_bits = binary_reordered.values.flatten().tolist()
        decoded_string = binary_labels_to_string(all_bits)
        st.subheader("Decoded String (continuous across rows)")
        st.write(decoded_string)

        st.subheader("Binary Labels (Ascending 3244→4882)")
        st.dataframe(binary_part.style.applymap(color_binary))
        st.download_button("Download Ascending Order CSV", binary_part.to_csv(index=False), "ef_binary_labels_ascending.csv", key="download_csv_tab4_ascend")

        all_bits = binary_part.values.flatten().tolist()
        decoded_string = binary_labels_to_string(all_bits)
        st.subheader("Decoded String (continuous across rows)")
        st.write(decoded_string)

# Tab 5: Binary → String
with tab5:
    st.header("Decode Binary Labels to String")

    st.subheader("🔘 Option 1: 32-bit Binary per Row")
    st.write("Upload CSV with 32 columns (0 or 1), no headers, from EF Binary format.")
    binary32_file = st.file_uploader("Upload 32-bit Binary CSV", type=["csv"], key="binary_32")

    if binary32_file:
        df_32 = pd.read_csv(binary32_file, header=None)
        if df_32.shape[1] != 32:
            st.warning("⚠️ CSV must have exactly 32 columns.")
        else:
            # Reordered: 4402 → 3244, 4882 → 4455
            df_32.columns = [str(h) for h in mutation_site_headers_3614]
            all_bits = df_32.values.flatten().astype(int).tolist()
            decoded_reordered = binary_labels_to_string(all_bits)

            st.subheader("Decoded String (Reordered 4402→3244, 4882→4455)")
            st.write(decoded_reordered)
            st.download_button("Download Reordered CSV", df_32.to_csv(index=False), "decoded_binary_32_reordered.csv", key="download_csv_tab5_32_reordered")

            # Ascending: 3244 → 4882
            df_ascending = df_32[[str(h) for h in mutation_site_headers_actual_3614 if str(h) in df_32.columns]]
            decoded_asc = binary_labels_to_string(df_ascending.values.flatten().astype(int).tolist())

            st.subheader("Decoded String (Ascending 3244→4882)")
            st.write(decoded_asc)
            st.download_button("Download Ascending CSV", df_ascending.to_csv(index=False), "decoded_binary_32_ascending.csv", key="download_csv_tab5_32_ascend")

    st.markdown("---")

    st.subheader("🔘 Option 2: 31-bit Binary Grouped per Row")
    st.write("Upload CSV with 31 columns per row (each row = one 6-bit ASCII chunk group).")
    binary31_file = st.file_uploader("Upload 31-bit Group CSV", type=["csv"], key="binary_31")

    if binary31_file:
        df_31 = pd.read_csv(binary31_file, header=None)

        def trim_row(row):
            bits = row.dropna().astype(int).tolist()[:31]
            return bits

        decoded_rows = []
        for _, row in df_31.iterrows():
            bits = trim_row(row)
            decoded_rows.append(binary_labels_to_string(bits))

        st.subheader("Decoded String from 31-bit Chunks")
        full_decoded = "".join(decoded_rows)
        st.write(full_decoded)

        st.download_button("Download Concatenated Output", full_decoded, "decoded_31bit_string.txt", key="download_csv_tab5_31")