Minor bug fixes

kiwijuice56 · Jan 17, 2024 · 291191a · 291191a
1 parent 64c9391
commit 291191a
Show file tree

Hide file tree

Showing 6 changed files with 19 additions and 8 deletions.
diff --git a/icon.ico → docs/favicon.ico b/icon.ico → docs/favicon.ico
diff --git a/docs/index.html b/docs/index.html
@@ -3,6 +3,7 @@
 <head>
     <title>protein-visualizer</title>
     <link rel="stylesheet" href="styles.css">
+    <link rel="icon" href="favicon.ico">
     <meta name="viewport" content="width=device-width, initial-scale=1">
 </head>
 <body>

diff --git a/img/icon.ico b/img/icon.ico
diff --git a/icon.png → img/icon.png b/icon.png → img/icon.png
diff --git a/main.py b/main.py
@@ -12,11 +12,15 @@
 from ctypes import windll
 
 
+print("protein-visualizer version 0.0, documentation: https://kiwijuice56.github.io/protein-visualizer/")
+print("-" * 8)
+print("Select a .pdb file to render.")
+
 # Fix screen resolution
 windll.shcore.SetProcessDpiAwareness(1)
 
 root = tk.Tk()
-root.iconbitmap("icon.ico")
+root.iconbitmap("img/icon.ico")
 root.withdraw()
 
 # Initialize protein
@@ -25,8 +29,8 @@
 # Initialize window
 window = pyglet.window.Window(resizable=True)
 window.set_caption("protein-visualizer")
-icon_ico = pyglet.image.load("icon.ico")
-icon_png = pyglet.image.load("icon.png")
+icon_ico = pyglet.image.load("img/icon.ico")
+icon_png = pyglet.image.load("img/icon.png")
 window.set_icon(icon_ico, icon_png)
 
 root.destroy()

diff --git a/protein.py b/protein.py
@@ -66,8 +66,11 @@ def __init__(self, pdb_path, chain_id=None, color_mode=CLUSTER_INDEX, color_pale
         self.color_mode = color_mode
         self.color_palette = color_palette
 
+        warnings.filterwarnings("ignore")
+
+        print("Loading protein structure.")
+
         # Get the full 3D structure from the PDB file
-        warnings.filterwarnings("ignore", "'HEADER' line not found; can't determine PDB ID.")
         bio_structure = Bio.PDB.PDBParser(QUIET=True).get_structure("struct", pdb_path)
 
         # Choose default chain if none specified
@@ -87,6 +90,8 @@ def __init__(self, pdb_path, chain_id=None, color_mode=CLUSTER_INDEX, color_pale
                 f"Chain with ID '{chain_id}' could not be found within .pdb file."
                 f"Available chains: {[chain.get_id() for chain in chains]}.")
 
+        print(f"Rendering chain with ID '{chain_id}' from {[chain.get_id() for chain in chains]}")
+
         # There are two methods of attaining an amino acid sequence from a PDB file: `atom` and `seqres`
         # `atom` calculates the sequence from the available residues found in the 3D space
         # `seqres` retrieves the sequence from the PDB file metadata
@@ -101,8 +106,7 @@ def __init__(self, pdb_path, chain_id=None, color_mode=CLUSTER_INDEX, color_pale
             if other_record.annotations["chain"] == chain_id:
                 full_record = other_record
         if full_record is None:
-            warnings.warn("Could not read SEQRES information; Continuing with sequence as determined from the 3D "
-                          "structure.", stacklevel=2)
+            print("Could not read SEQRES info; Continuing with sequence as determined from the 3D structure.")
             full_record = physical_record
 
         prot_name = "".join(x for x in full_record.name if x.isalnum())
@@ -135,11 +139,11 @@ def __init__(self, pdb_path, chain_id=None, color_mode=CLUSTER_INDEX, color_pale
 
         # Calculate embeddings with ProSE
         if not os.path.isfile(f"data/{prot_name}_{chain_id}_embeddings.h5"):
-            print("Embeddings not found in `data` directory. Generating new embeddings.")
+            print("Embeddings not found in 'data' directory. Generating new embeddings.")
             subprocess.call(["python", "prose/embed_sequences.py", '-o', f"data/{prot_name}_{chain_id}_embeddings.h5",
                              f"data/{prot_name}_{chain_id}.fa"], shell=True)
         else:
-            print("Embeddings from previous session found in `data` directory.")
+            print("Embeddings from previous session found in 'data' directory.")
 
         # Parse the embeddings file for the residues we will render
         embedding_file = h5py.File(f"data/{prot_name}_{chain_id}_embeddings.h5", 'r')
@@ -148,6 +152,8 @@ def __init__(self, pdb_path, chain_id=None, color_mode=CLUSTER_INDEX, color_pale
         for residue in self.residues:
             realized_embeddings.append(embeddings[residue.seq_index - seq_index_offset])
 
+        print("Calculating 2D projection and clusters.")
+
         # Use the t-SNE algorithm to transform the embeddings into 2D vectors
         transform = TSNE(n_components=2, perplexity=3).fit_transform(np.array(realized_embeddings))
         self.embedding_points = transform.flatten()