Upload app.py

app.py

@@ -0,0 +1,174 @@
+import gradio as gr
+import numpy as np
+import random
+import pandas as pd
+import matplotlib.pyplot as plt
+from io import BytesIO
+import base64
+
+# 模拟数据 - 实际使用时需要替换为真实数据
+species_data = {
+    "human": {"codon_table": {}, "trna": {}, "codon_usage": {}},
+    "mouse": {"codon_table": {}, "trna": {}, "codon_usage": {}},
+    "virus": {"codon_table": {}, "trna": {}, "codon_usage": {}},
+    "Escherichia coli": {"codon_table": {}, "trna": {}, "codon_usage": {}},
+    "酿酒酵母": {"codon_table": {}, "trna": {}, "codon_usage": {}},
+    "Pichia": {"codon_table": {}, "trna": {}, "codon_usage": {}},
+}
+
+# 模拟函数 - 实际需要生物信息学算法实现
+def find_longest_cds(seq):
+    # 简化的ORF查找 - 实际应使用生物信息学库
+    start = seq.find("ATG")
+    stops = [seq.find("TAA", start), seq.find("TAG", start), seq.find("TGA", start)]
+    stops = [s for s in stops if s > start]
+    end = min(stops) + 3 if stops else len(seq)
+    return start, end
+
+def calculate_cds_variants(protein_seq):
+    # 简化的计算 - 实际应根据密码子表计算
+    aa_count = len(protein_seq)
+    return 2 ** aa_count  # 示例值
+
+def optimize_cds(protein_seq, species, method):
+    # 生成20个优化序列示例
+    results = []
+    for i in range(20):
+        # 实际应根据优化方法生成序列
+        seq = ''.join(random.choices("ACGT", k=len(protein_seq)*3))
+        gc = random.uniform(0.3, 0.7)
+        trna = random.uniform(0.5, 1.0)
+        usage = random.uniform(0.6, 0.95)
+        mfe = random.uniform(-30, -10)
+        score = gc*0.25 + trna*0.25 + usage*0.25 + (-mfe/40)*0.25
+        
+        results.append({
+            "Sequence": seq,
+            "GC%": f"{gc*100:.1f}%",
+            "tRNA": f"{trna:.3f}",
+            "Usage": f"{usage:.3f}",
+            "MFE": f"{mfe:.1f}",
+            "Score": f"{score:.3f}"
+        })
+    return pd.DataFrame(results)
+
+def design_mrna(utr5_candidates, utr3_candidates, cds_seq):
+    # 生成20个设计结果示例
+    designs = []
+    for i in range(20):
+        utr5 = random.choice(utr5_candidates)
+        utr3 = random.choice(utr3_candidates)
+        full_seq = utr5 + cds_seq + utr3
+        mfe = random.uniform(-50, -20)
+        designs.append({
+            "Design": f"Design_{i+1}",
+            "5'UTR": utr5[:10] + "..." if len(utr5) > 13 else utr5,
+            "3'UTR": utr3[:10] + "..." if len(utr3) > 13 else utr3,
+            "MFE": f"{mfe:.1f}",
+            "Sequence": full_seq
+        })
+    return pd.DataFrame(designs)
+
+# 标注可视化函数
+def visualize_annotation(seq):
+    start, end = find_longest_cds(seq)
+    html = f"""
+    <div style="font-family: monospace; font-size: 14px; line-height: 1.8;">
+        <div style="background-color: #ffcccc; display: inline-block; padding: 2px;">
+            5'UTR: {seq[:start] if start > 0 else 'N/A'}
+        </div>
+        <div style="background-color: #ccffcc; display: inline-block; padding: 2px;">
+            CDS: {seq[start:end] if start >=0 else 'N/A'}
+        </div>
+        <div style="background-color: #ccccff; display: inline-block; padding: 2px;">
+            3'UTR: {seq[end:] if end < len(seq) else 'N/A'}
+        </div>
+    </div>
+    <p>Annotation Legend:</p>
+    <div style="display: flex; gap: 10px;">
+        <div style="background-color: #ffcccc; padding: 5px;">5'UTR</div>
+        <div style="background-color: #ccffcc; padding: 5px;">CDS</div>
+        <div style="background-color: #ccccff; padding: 5px;">3'UTR</div>
+    </div>
+    """
+    return html
+
+# 创建Gradio界面
+with gr.Blocks(title="Vaccine Designer", theme=gr.themes.Soft()) as app:
+    gr.Markdown("# 🧬 Vaccine Design Platform - Academic Collaboration")
+    
+    with gr.Tab("mRNA Annotation"):
+        gr.Markdown("## mRNA Sequence Annotation")
+        mrna_input = gr.Textbox(label="mRNA Sequence", placeholder="Enter mRNA sequence here...")
+        annotate_btn = gr.Button("Annotate Regions")
+        annotation_output = gr.HTML(label="Sequence Annotation")
+        annotate_btn.click(visualize_annotation, inputs=mrna_input, outputs=annotation_output)
+    
+    with gr.Tab("CDS Variants"):
+        gr.Markdown("## Calculate Potential CDS Variants")
+        protein_input = gr.Textbox(label="Protein Sequence", placeholder="Enter protein sequence here...")
+        calc_btn = gr.Button("Calculate Variants")
+        variants_output = gr.Number(label="Potential CDS Variants")
+        calc_btn.click(calculate_cds_variants, inputs=protein_input, outputs=variants_output)
+    
+    with gr.Tab("CDS Optimization"):
+        gr.Markdown("## Optimize CDS Sequence")
+        with gr.Row():
+            protein_seq = gr.Textbox(label="Protein Sequence")
+            species = gr.Dropdown(list(species_data.keys()), label="Species", value="human")
+        
+        method = gr.Radio(["Max GC Content", "tRNA Abundance", "Codon Usage", "MFE Optimization"], 
+                          label="Optimization Method", value="Max GC Content")
+        
+        optimize_btn = gr.Button("Generate Optimized Sequences")
+        results_table = gr.Dataframe(label="Top 20 Optimized Sequences", headers=["Sequence", "GC%", "tRNA", "Usage", "MFE", "Score"])
+        optimize_btn.click(optimize_cds, inputs=[protein_seq, species, method], outputs=results_table)
+        
+        # 评分可视化
+        plot = gr.Plot(label="Optimization Scores")
+        def update_plot(df):
+            if df is None or len(df) == 0:
+                return None
+            fig, ax = plt.subplots()
+            scores = [float(x) for x in df["Score"]]
+            ax.bar(range(len(scores)), scores, color='skyblue')
+            ax.set_xlabel("Sequence Rank")
+            ax.set_ylabel("Composite Score")
+            ax.set_title("Optimization Scores of Top Sequences")
+            return fig
+        results_table.change(update_plot, inputs=results_table, outputs=plot)
+    
+    with gr.Tab("Full mRNA Design"):
+        gr.Markdown("## Design Full mRNA Sequence")
+        with gr.Row():
+            utr5_upload = gr.File(label="Upload 5'UTR Candidates (txt)")
+            utr3_upload = gr.File(label="Upload 3'UTR Candidates (txt)")
+        cds_input = gr.Textbox(label="CDS Sequence")
+        design_btn = gr.Button("Design mRNA Sequences")
+        design_results = gr.Dataframe(label="Top 20 Designs", headers=["Design", "5'UTR", "3'UTR", "MFE", "Sequence"])
+        design_btn.click(design_mrna, inputs=[utr5_upload, utr3_upload, cds_input], outputs=design_results)
+    
+    with gr.Tab("Resources & Links"):
+        gr.Markdown("## Helpful Resources")
+        gr.Markdown("""
+        - [mRNA Designer Platform](https://www.biosino.org/mRNAdesigner/main)
+        - [Nucleic Acid Database](https://ngdc.cncb.ac.cn/ncov/)
+        - [NCBI GenBank](https://www.ncbi.nlm.nih.gov/genbank/)
+        - [ViralZone](https://viralzone.expasy.org/)
+        - [Codon Usage Database](https://www.kazusa.or.jp/codon/)
+        """)
+        
+        gr.Markdown("## Download All Results")
+        download_btn = gr.Button("Download Results Package")
+        download_btn.click(lambda: "results.zip", outputs=gr.File(label="Download Results"))
+    
+    gr.Markdown("---")
+    gr.HTML("""
+    <div style="text-align: center; padding: 20px; background-color: #f0f0f0; border-radius: 10px;">
+        <p>Academic Collaboration Platform for Vaccine Design | Contact: [email protected]</p>
+    </div>
+    """)
+
+# 运行应用
+if __name__ == "__main__":
+    app.launch(server_name="0.0.0.0", server_port=7860)