diff --git a/concha/utils.ipynb b/concha/utils.ipynb
index c672672bf682d2af877f647b4aa5da01b4fe9542..adae05ba13e9a7a33f53e604512657394e24cc9d 100644
--- a/concha/utils.ipynb
+++ b/concha/utils.ipynb
@@ -25,7 +25,7 @@
"metadata": {},
"outputs": [],
"source": [
- "def get_cluster_files(ensemble_path, ensemble_name, clustering_partition):\n",
+ "def get_cluster_files(ensemble_path, ensemble_name, labels_umap):\n",
" \n",
" # Initial parameters\n",
" var_dict = {'multiframe' : 'n', 'check_folder' : True, 'do_xtc' : False, 'do_pdb' : False,\n",
@@ -94,19 +94,19 @@
" traj_file = md.load_xtc(\"/\".join([var_dict[\"xtc_root_path\"],var_dict[\"xtc_files\"][0]]), top = \"/\".join([var_dict[\"xtc_root_path\"],var_dict[\"pdb_files\"][0]]))\n",
" \n",
" # Save .xtc cluster files\n",
- " for k in tqdm(range(len(np.unique(clustering_partition[clustering_partition >= 0])))):\n",
- " traj_file[np.where(clustering_partition == k)].save_xtc(\"/\".join([save_files, \"\".join([ensemble_name,'_',str(k),'.xtc'])]))\n",
+ " for k in tqdm(range(len(np.unique(labels_umap[labels_umap >= 0])))):\n",
+ " traj_file[np.where(labels_umap == k)].save_xtc(\"/\".join([save_files, \"\".join([ensemble_name,'_',str(k),'.xtc'])]))\n",
"\n",
" if var_dict[\"do_pdb\"]:\n",
" \n",
" conf_list = os.listdir(\"/\".join([var_dict[\"ensemble_path\"],var_dict[\"folders\"][0]]))\n",
"\n",
- " for k in tqdm(range(len(np.unique(clustering_partition[clustering_partition >= 0])))):\n",
+ " for k in tqdm(range(len(np.unique(labels_umap[labels_umap >= 0])))):\n",
" clus_k_path = \"/\".join([save_path, \"_\".join(['clus',str(k)])])\n",
" if not os.path.exists(clus_k_path):\n",
" os.mkdir(clus_k_path)\n",
" \n",
- " clus_k = np.where(clustering_partition == k)[0]\n",
+ " clus_k = np.where(labels_umap == k)[0]\n",
" for j in range(len(clus_k)):\n",
" traj = md.load_pdb(\"/\".join([\"/\".join([var_dict[\"ensemble_path\"],var_dict[\"folders\"][0]]),conf_list[clus_k[j]]]))\n",
" traj.save_pdb(\"/\".join([clus_k_path, \"\".join([ensemble_name,'_',str(clus_k[j]),'.pdb'])]))\n",
@@ -121,9 +121,9 @@
"metadata": {},
"outputs": [],
"source": [
- "def plot_2umap(embedding_2d, clustering_partition, ensemble_name, results_path):\n",
+ "def plot_2umap(embedding_2d, labels_umap, ensemble_name, results_path):\n",
" \n",
- " classified = np.where(clustering_partition >= 0)[0]\n",
+ " classified = np.where(labels_umap >= 0)[0]\n",
" \n",
" output1 = widgets.Output()\n",
" with output1:\n",
@@ -135,7 +135,7 @@
" alpha=0.5)\n",
" scatter = ax.scatter(embedding_2d[classified, 0],\n",
" embedding_2d[classified, 1],\n",
- " c=clustering_partition[classified],\n",
+ " c=labels_umap[classified],\n",
" s=0.5,\n",
" alpha = 1,\n",
" cmap='Spectral')\n",
@@ -147,9 +147,9 @@
"\n",
" output2 = widgets.Output()\n",
" with output2:\n",
- " repartition = pd.Series(clustering_partition).value_counts()\n",
+ " repartition = pd.Series(labels_umap).value_counts()\n",
" repartition.index = [\"Unclassified\" if i == -1 else i for i in repartition.index]\n",
- " display(pd.DataFrame({\"Cluster\" : np.array(repartition.index), \"Occupancy (%)\" : 100*np.array(repartition.values)/len(clustering_partition)}))\n",
+ " display(pd.DataFrame({\"Cluster\" : np.array(repartition.index), \"Occupancy (%)\" : 100*np.array(repartition.values)/len(labels_umap)}))\n",
" two_columns = widgets.HBox([output1, output2])\n",
" display(two_columns)"
]
@@ -161,7 +161,7 @@
"metadata": {},
"outputs": [],
"source": [
- "def get_wmaps(wcont_data, clustering_partition, ensemble_name, results_path):\n",
+ "def get_wmaps(wcont_data, labels_umap, ensemble_name, results_path):\n",
" \n",
" \n",
" maps_path = \"/\".join([results_path,\"wcont_maps\"]) # Path to save files\n",
@@ -203,7 +203,7 @@
"metadata": {},
"outputs": [],
"source": [
- "def representative_ensemble(size, ensemble_path, ensemble_name, clustering_partition):\n",
+ "def representative_ensemble(size, ensemble_path, ensemble_name, labels_umap):\n",
" \n",
" # Initial parameters\n",
" var_dict = {'multiframe' : 'n', 'check_folder' : True, 'do_xtc' : False, 'do_pdb' : False,\n",
@@ -262,7 +262,7 @@
" print(\"\\n----------------------------------------------------------------------------------\\n\")\n",
" print(\"\\nSampling representative family...\\n\")\n",
" \n",
- " repartition = pd.Series(clustering_partition).value_counts() # Clustering partition\n",
+ " repartition = pd.Series(labels_umap).value_counts() # Clustering partition\n",
" repartition.index = [\"Unclassified\" if i == -1 else i for i in repartition.index]\n",
" repartition = repartition.drop(\"Unclassified\")\n",
" probas = repartition.values/np.sum(repartition.values)\n",
@@ -271,7 +271,7 @@
" for i in range(size):\n",
"\n",
" choose_cluster = np.random.choice(repartition.index, size = 1, p = probas)[0]\n",
- " selected_conf[i] = np.random.choice(np.where(clustering_partition == choose_cluster)[0], size = 1)[0]\n",
+ " selected_conf[i] = np.random.choice(np.where(labels_umap == choose_cluster)[0], size = 1)[0]\n",
" \n",
" selected_conf = np.ndarray.astype(selected_conf, int)\n",
" results_path = \"/\".join([os.path.abspath(ensemble_path),\"_\".join(['results',ensemble_name])])\n",
@@ -303,7 +303,7 @@
"metadata": {},
"outputs": [],
"source": [
- "def cluster_descriptors(ensemble_path, ensemble_name, clustering_partition):\n",
+ "def cluster_descriptors(ensemble_path, ensemble_name, labels_umap):\n",
" \n",
" # Initial parameters\n",
" var_dict = {'multiframe' : 'n', 'check_folder' : True, 'do_xtc' : False, 'do_pdb' : False,\n",
@@ -374,16 +374,16 @@
" Nconf = traj_file.n_frames\n",
" \n",
" dssp_types = ['H','B','E','G','I','T','S',' ']\n",
- " prop_dssp = np.zeros([len(dssp_types),L,len(clustering_partition)-1])\n",
- " rg = np.zeros([len(clustering_partition)-1])\n",
+ " prop_dssp = np.zeros([len(dssp_types),L,len(labels_umap)-1])\n",
+ " rg = np.zeros([len(labels_umap)-1])\n",
" \n",
- " for k in range(len(np.unique(clustering_partition[clustering_partition >= 0]))):\n",
+ " for k in range(len(np.unique(labels_umap[labels_umap >= 0]))):\n",
" \n",
" prop_dssp_k = np.zeros([len(dssp_types),L])\n",
- " dssp_k = md.compute_dssp(traj_file[np.where(clustering_partition == k)], simplified = False)\n",
- " rg[k] = np.mean(md.compute_rg(traj_file[np.where(clustering_partition == k)]))\n",
+ " dssp_k = md.compute_dssp(traj_file[np.where(labels_umap == k)], simplified = False)\n",
+ " rg[k] = np.mean(md.compute_rg(traj_file[np.where(labels_umap == k)]))\n",
" for dt in range(len(dssp_types)):\n",
- " prop_dssp_k[dt,:] = (dssp_k == dssp_types[dt]).sum(axis = 0)/len(np.where(clustering_partition == k)[0])\n",
+ " prop_dssp_k[dt,:] = (dssp_k == dssp_types[dt]).sum(axis = 0)/len(np.where(labels_umap == k)[0])\n",
" prop_dssp[:,:,k] = prop_dssp_k\n",
"\n",
" if var_dict[\"do_pdb\"]:\n",
@@ -394,13 +394,13 @@
" Nconf = len(conf_list)\n",
" \n",
" dssp_types = ['H','B','E','G','I','T','S',' ']\n",
- " prop_dssp = np.zeros([len(dssp_types),L,len(clustering_partition)-1])\n",
- " rg = np.zeros([len(clustering_partition)-1])\n",
+ " prop_dssp = np.zeros([len(dssp_types),L,len(labels_umap)-1])\n",
+ " rg = np.zeros([len(labels_umap)-1])\n",
" \n",
- " for k in range(len(np.unique(clustering_partition[clustering_partition >= 0]))):\n",
+ " for k in range(len(np.unique(labels_umap[labels_umap >= 0]))):\n",
" \n",
" prop_dssp_k = np.zeros([len(dssp_types),L])\n",
- " clus_k = np.where(clustering_partition == k)[0]\n",
+ " clus_k = np.where(labels_umap == k)[0]\n",
" dssp_k = np.zeros([len(clus_k),L]).astype(str)\n",
" rg_k = np.zeros([len(clus_k)])\n",
"\n",
@@ -409,12 +409,12 @@
" rg_k[l] = md.compute_rg(md.load_pdb(\"/\".join([pdb_folder,conf_list[clus_k[l]]])))\n",
" rg[k] = np.mean(rg_k)\n",
" for dt in range(len(dssp_types)):\n",
- " prop_dssp_k[dt,:] = (dssp_k == dssp_types[dt]).sum(axis = 0)/len(np.where(clustering_partition == k)[0])\n",
+ " prop_dssp_k[dt,:] = (dssp_k == dssp_types[dt]).sum(axis = 0)/len(np.where(labels_umap == k)[0])\n",
" prop_dssp[:,:,k] = prop_dssp_k\n",
" \n",
- " for cluster in tqdm(range(len(np.unique(clustering_partition[clustering_partition >= 0])))):\n",
+ " for cluster in tqdm(range(len(np.unique(labels_umap[labels_umap >= 0])))):\n",
" \n",
- " prop_cluster = round(100*len(np.where(clustering_partition == cluster)[0])/Nconf,2)\n",
+ " prop_cluster = round(100*len(np.where(labels_umap == cluster)[0])/Nconf,2)\n",
" fig = plt.figure(figsize=(10, 1.7))\n",
" res = sns.heatmap(prop_dssp[:,:,cluster], cmap='Blues', square = True, cbar_kws={\"shrink\": .7,'label':\"Class prop.\"})\n",
" xlabels = [item.get_text() for item in res.get_xmajorticklabels()]\n",