re-type commited on
Commit
824ed48
·
verified ·
1 Parent(s): 419d8d9

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +17 -17
app.py CHANGED
@@ -6,7 +6,7 @@ import subprocess
6
  import pandas as pd
7
  from predictor import GenePredictor
8
  from tensorflow.keras.models import load_model
9
- from ml_simplified_tree import maximum_likelihood
10
  import logging
11
  import os
12
 
@@ -19,9 +19,9 @@ try:
19
  keras_model = load_model("best_model.keras")
20
  with open("kmer_to_index.pkl", "rb") as f:
21
  kmer_to_index = pickle.load(f)
22
- logging.info("Models and kmer_to_index loaded successfully")
23
  except FileNotFoundError as e:
24
- logging.error(f"Model or file not found: {e}")
25
  raise
26
  except Exception as e:
27
  logging.error(f"Error loading models or files: {e}")
@@ -51,7 +51,7 @@ def save_to_fasta(name, sequence, path):
51
 
52
  def save_to_csv(sequence, path):
53
  try:
54
- df = pd.DataFrame({"Sequence": [sequence]})
55
  df.to_csv(path, index=False)
56
  logging.info(f"CSV file saved to {path}")
57
  return "CSV file saved successfully"
@@ -93,19 +93,19 @@ def run_full_pipeline(dna_input):
93
  # Step 3b: Run MAFFT and IQTree
94
  mafft_status = run_mafft_and_iqtree(step2_out)
95
 
96
- # Step 3c: Run ML Simplified Tree
97
- html_file = "tree.html"
98
- try:
99
- ml_output = maximum_likelihood("f_gene_clean_dataset.csv")
100
- if os.path.exists(html_file):
101
- logging.info(f"HTML tree file generated: {html_file}")
 
102
  else:
103
- logging.warning(f"HTML tree file {html_file} not found")
104
- html_file = None
105
- except Exception as e:
106
- logging.error(f"ML Tree Error: {e}")
107
- ml_output = f"ML Tree Error: {e}"
108
- html_file = None
109
 
110
  return {
111
  "Boundary Model Output": step1_out,
@@ -131,7 +131,7 @@ def run_full_pipeline(dna_input):
131
  # --------- Gradio Interface and API ---------
132
  with gr.Blocks() as gr_interface:
133
  gr.Markdown("# Sequential Phylogenetic Inference Pipeline")
134
- gr.Markdown("This pipeline runs sequentially: Boundary-Aware Model → Keras Model → MAFFT/IQTree & ML Tree")
135
 
136
  dna_input = gr.Textbox(label="Input DNA Sequence")
137
  submit_button = gr.Button("Run Pipeline")
 
6
  import pandas as pd
7
  from predictor import GenePredictor
8
  from tensorflow.keras.models import load_model
9
+ import ml_simplified_tree # Import the module
10
  import logging
11
  import os
12
 
 
19
  keras_model = load_model("best_model.keras")
20
  with open("kmer_to_index.pkl", "rb") as f:
21
  kmer_to_index = pickle.load(f)
22
+ logging.info("Models and kmer_to_index loaded successfully from Space")
23
  except FileNotFoundError as e:
24
+ logging.error(f"Model or file not found in Space: {e}")
25
  raise
26
  except Exception as e:
27
  logging.error(f"Error loading models or files: {e}")
 
51
 
52
  def save_to_csv(sequence, path):
53
  try:
54
+ df = pd.DataFrame({"F-gene": [sequence], "Accession Number": ["Query"], "ML": ["Unknown"], "Genotype": ["Unknown"]})
55
  df.to_csv(path, index=False)
56
  logging.info(f"CSV file saved to {path}")
57
  return "CSV file saved successfully"
 
93
  # Step 3b: Run MAFFT and IQTree
94
  mafft_status = run_mafft_and_iqtree(step2_out)
95
 
96
+ # Step 3c: Run ML Simplified Tree using PhylogeneticTreeAnalyzer
97
+ analyzer = ml_simplified_tree.PhylogeneticTreeAnalyzer()
98
+ if not analyzer.load_data("f_gene_clean_dataset.csv"):
99
+ ml_output = "Error: Failed to load data"
100
+ else:
101
+ if not analyzer.find_query_sequence(step2_out):
102
+ ml_output = "Error: Invalid query sequence"
103
  else:
104
+ matched_ids, actual_percentage = analyzer.find_similar_sequences(analyzer.matching_percentage)
105
+ analyzer.create_interactive_tree(matched_ids, actual_percentage)
106
+ ml_output = f"Tree generated and saved as phylogenetic_tree_normalized_horizontal.html" if os.path.exists("phylogenetic_tree_normalized_horizontal.html") else "Error: Tree generation failed"
107
+
108
+ html_file = "phylogenetic_tree_normalized_horizontal.html" if os.path.exists("phylogenetic_tree_normalized_horizontal.html") else None
 
109
 
110
  return {
111
  "Boundary Model Output": step1_out,
 
131
  # --------- Gradio Interface and API ---------
132
  with gr.Blocks() as gr_interface:
133
  gr.Markdown("# Sequential Phylogenetic Inference Pipeline")
134
+ gr.Markdown("This pipeline runs: Boundary-Aware Model → Keras Model → MAFFT/IQTree & ML Tree")
135
 
136
  dna_input = gr.Textbox(label="Input DNA Sequence")
137
  submit_button = gr.Button("Run Pipeline")