lint

Author: berombau

docs/notebooks/example.ipynb

@@ -644,7 +644,7 @@
    ],
    "source": [
     "# Visualize data\n",
-    "ax = plt.hist2d(ff_t[:,\"CD4\"].X.flatten(), ff_t[:,\"CD8\"].X.flatten(), bins=200, cmin = 1, cmap = \"jet\")\n",
+    "ax = plt.hist2d(ff_t[:, \"CD4\"].X.flatten(), ff_t[:, \"CD8\"].X.flatten(), bins=200, cmin=1, cmap=\"jet\")\n",
     "plt.show()"
    ]
   },
@@ -714,7 +714,7 @@
     }
    ],
    "source": [
-    "fsom = fs.FlowSOM(ff_t, cols_to_use, xdim = 10, ydim = 10, n_clus = 10)\n",
+    "fsom = fs.FlowSOM(ff_t, cols_to_use, xdim=10, ydim=10, n_clus=10)\n",
     "fsom.mudata"
    ]
   },
@@ -749,7 +749,7 @@
     }
    ],
    "source": [
-    "ff_clustered = fs.flowsom_clustering(ff_t, cols_to_use, xdim = 10, ydim = 10, n_clus = 10)\n",
+    "ff_clustered = fs.flowsom_clustering(ff_t, cols_to_use, xdim=10, ydim=10, n_clus=10)\n",
     "ff_clustered"
    ]
   },
@@ -857,7 +857,7 @@
     }
    ],
    "source": [
-    "p = fs.pl.plot_stars(fsom, background_values = fsom.get_cluster_data().obs.metaclustering)\n",
+    "p = fs.pl.plot_stars(fsom, background_values=fsom.get_cluster_data().obs.metaclustering)\n",
     "p.show()"
    ]
   },
@@ -897,7 +897,13 @@
     }
    ],
    "source": [
-    "p = fs.pl.plot_stars(fsom, background_values = fsom.get_cluster_data().obs.metaclustering, view = \"grid\", equal_node_size = True, equal_background_size = True)\n",
+    "p = fs.pl.plot_stars(\n",
+    "    fsom,\n",
+    "    background_values=fsom.get_cluster_data().obs.metaclustering,\n",
+    "    view=\"grid\",\n",
+    "    equal_node_size=True,\n",
+    "    equal_background_size=True,\n",
+    ")\n",
     "p"
    ]
   },
@@ -927,12 +933,12 @@
    ],
    "source": [
     "# Read in that data\n",
-    "file = open(\"../../tests/data/gating_result.csv\", \"r\")\n",
+    "file = open(\"../../tests/data/gating_result.csv\")\n",
     "data = csv.reader(file)\n",
     "data = [i[0] for i in data]\n",
     "\n",
     "# Plot\n",
-    "p = fs.pl.plot_pies(fsom, data, background_values = fsom.get_cluster_data().obs.metaclustering)"
+    "p = fs.pl.plot_pies(fsom, data, background_values=fsom.get_cluster_data().obs.metaclustering)"
    ]
   },
   {
@@ -962,7 +968,7 @@
     }
    ],
    "source": [
-    "p = fs.pl.plot_numbers(fsom, level = \"clusters\", text_size = 5)"
+    "p = fs.pl.plot_numbers(fsom, level=\"clusters\", text_size=5)"
    ]
   },
   {
@@ -991,7 +997,7 @@
     }
    ],
    "source": [
-    "p = fs.pl.plot_marker(fsom, marker = np.array([\"CD3\"]))"
+    "p = fs.pl.plot_marker(fsom, marker=np.array([\"CD3\"]))"
    ]
   },
   {
@@ -1028,13 +1034,13 @@
    ],
    "source": [
     "p = fs.pl.plot_2D_scatters(\n",
-    "        fsom,\n",
-    "        channelpairs=[[\"CD3\", \"CD4\"], [\"CD19\", \"TCRb\"]],\n",
-    "        clusters=[[1, 2], [3]],\n",
-    "        metaclusters=[[4], [5, 6]],\n",
-    "        density=True,\n",
-    "        centers=True,\n",
-    "    )\n",
+    "    fsom,\n",
+    "    channelpairs=[[\"CD3\", \"CD4\"], [\"CD19\", \"TCRb\"]],\n",
+    "    clusters=[[1, 2], [3]],\n",
+    "    metaclusters=[[4], [5, 6]],\n",
+    "    density=True,\n",
+    "    centers=True,\n",
+    ")\n",
     "p.show()"
    ]
   },
@@ -1148,7 +1154,12 @@
     }
    ],
    "source": [
-    "features = fs.tl.get_features(fsom, files=[ff_t[1:1000, :], ff_t[1000:2000, :]], level = [\"clusters\", \"metaclusters\"], type = [\"counts\", \"percentages\"])\n",
+    "features = fs.tl.get_features(\n",
+    "    fsom,\n",
+    "    files=[ff_t[1:1000, :], ff_t[1000:2000, :]],\n",
+    "    level=[\"clusters\", \"metaclusters\"],\n",
+    "    type=[\"counts\", \"percentages\"],\n",
+    ")\n",
     "features[\"metacluster_percentages\"]"
    ]
   },
@@ -1263,7 +1274,7 @@
     }
    ],
    "source": [
-    "fs.tl.get_counts(fsom, level = \"clusters\")"
+    "fs.tl.get_counts(fsom, level=\"clusters\")"
    ]
   },
   {
@@ -1433,7 +1444,9 @@
     }
    ],
    "source": [
-    "fs.pp.aggregate_flowframes(filenames = [\"../../tests/data/not_preprocessed.fcs\", \"../../tests/data/not_preprocessed.fcs\"], c_total = 5000)"
+    "fs.pp.aggregate_flowframes(\n",
+    "    filenames=[\"../../tests/data/not_preprocessed.fcs\", \"../../tests/data/not_preprocessed.fcs\"], c_total=5000\n",
+    ")"
    ]
   },
   {
@@ -1496,7 +1509,7 @@
    ],
    "source": [
     "fsom_new = fsom.new_data(ff_t[1:200, :])\n",
-    "fs.pl.plot_stars(fsom_new, background_values = fsom_new.get_cluster_data().obs.metaclustering)"
+    "fs.pl.plot_stars(fsom_new, background_values=fsom_new.get_cluster_data().obs.metaclustering)"
    ]
   },
   {
@@ -1545,7 +1558,7 @@
    ],
    "source": [
     "fsom_subset = fsom.subset(fsom.get_cell_data().obs[\"metaclustering\"] == 4)\n",
-    "fs.pl.plot_stars(fsom_subset, background_values = fsom_subset.get_cluster_data().obs.metaclustering)"
+    "fs.pl.plot_stars(fsom_subset, background_values=fsom_subset.get_cluster_data().obs.metaclustering)"
    ]
   }
  ],

src/FlowSOM/io/read_fcs.py

@@ -19,7 +19,7 @@ def read_FCS(filepath):
         f.var = f.var.sort_values(by="n")
         f.uns["meta"]["channels"].index = f.uns["meta"]["channels"].index.astype(int)
         f.uns["meta"]["channels"] = f.uns["meta"]["channels"].sort_index()
-    except:
+    except ValueError:
         f = pm.io.read_fcs(filepath, reindex=False)
         markers = {
             str(re.sub("S$", "", re.sub("^P", "", string))): f.uns["meta"][string]
@@ -37,6 +37,7 @@ def read_FCS(filepath):
 
 
 def read_csv(filepath, spillover=None, **kwargs):
+    """Reads in a CSV file."""
     ff = ad.read_csv(filepath, **kwargs)
     ff.var = pd.DataFrame(
         {"n": range(ff.shape[1]), "channel": ff.var_names, "marker": ff.var_names}, index=ff.var.index

src/FlowSOM/main.py

@@ -290,6 +290,19 @@ def metacluster(self, n_clus):
     def consensus_hierarchical_clustering(
         self, data, n_clus, n_subsamples=100, linkage="average", resample_proportion=0.9
     ):
+        """Perform a consensus hierarchical clustering.
+
+        :param data: The data to cluster
+        :type data: np.array
+        :param n_clus: The number of metaclusters
+        :type n_clus: int
+        :param n_subsamples: The number of subsamples to use for consensus clustering
+        :type n_subsamples: int
+        :param linkage: The linkage method to use for clustering
+        :type linkage: str
+        :param resample_proportion: The proportion of the data to use for subsampling
+        :type resample_proportion: float
+        """
         average = ConsensusCluster(
             AgglomerativeClustering, K=n_clus, H=n_subsamples, resample_proportion=resample_proportion, linkage=linkage
         )
@@ -439,15 +452,18 @@ def subset(self, ids):
         return fsom_subset
 
     def get_cell_data(self):
+        """Get the cell data."""
         return self.mudata["cell_data"]
 
     def get_cluster_data(self):
+        """Get the cluster data."""
         return self.mudata["cluster_data"]
 
 
 def flowsom_clustering(inp, cols_to_use=None, n_clus=10, xdim=10, ydim=10, seed=None, **kwargs):
-    """Perform FlowSOM clustering on an anndata object and returns the anndata
-       object with the FlowSOM clusters and metaclusters added as variable.
+    """Perform FlowSOM clustering on an anndata object and returns the anndata object.
+
+    The FlowSOM clusters and metaclusters are added as variable.
 
     :param inp: An anndata or filepath to an FCS file
     :type inp: ad.AnnData / str

src/FlowSOM/pl/__init__.py

@@ -1,2 +0,0 @@
-from ._plot_helper_functions import *
-from .plot_functions import *

src/FlowSOM/pl/plot_functions.py

@@ -14,7 +14,18 @@
 
 from flowsom.tl import get_channels, get_markers
 
-from ._plot_helper_functions import *
+from ._plot_helper_functions import (
+    FlowSOM_colors,
+    add_legend,
+    add_nodes,
+    add_stars,
+    add_text,
+    add_wedges,
+    gg_color_hue,
+    plot_FlowSOM,
+    plot_star_legend,
+    scale_star_heights,
+)
 
 
 def plot_2D_scatters(
@@ -195,7 +206,8 @@ def plot_labels(fsom, labels, max_node_size=0, text_size=20, text_color="black",
 
 
 def plot_numbers(fsom, level="clusters", max_node_size=0, **kwargs):
-    """Plot cluster ids for each cluster
+    """Plot cluster ids for each cluster.
+
     :param fsom: A FlowSOM object
     :type fsom: FlowSOM
     :param level: Should be either "clusters" (default) or "metaclusters".
@@ -339,8 +351,7 @@ def plot_pies(
     title=None,
     **kwargs,
 ):
-    """Plot FlowSOM grid or tree, with pies indicating another clustering or
-    manual gating result.
+    """Plot FlowSOM grid or tree, with pies indicating another clustering or manual gating result.
 
     :param fsom: A FlowSOM object
     :type fsom: FlowSOM
@@ -467,7 +478,6 @@ def FlowSOMmary(fsom, plot_file="./FlowSOMmary.pdf"):
         np.random.choice(range(fsom.get_cell_data().shape[0]), 5000, replace=False),
         fsom.get_cell_data().var_names[ref_markers_bool],
     ]
-    subset_fsom
     neighbors(subset_fsom)
     umap(subset_fsom)
 

src/FlowSOM/pp/fcs_functions.py

@@ -12,7 +12,8 @@
 
 
 def aggregate_flowframes(filenames, c_total, channels=None, keep_order=False):
-    """Aggregate multiple FCS files together
+    """Aggregate multiple FCS files together.
+
     :param filenames: An array containing full paths to the FCS files
     :type filenames: np.array
     :param c_total: Total number of cells to write to the output file
@@ -56,6 +57,17 @@ def aggregate_flowframes(filenames, c_total, channels=None, keep_order=False):
 
 
 def normalize_estimate_logicle(adata, channels, m=4.5, q=0.05):
+    """Normalize and estimate logicle parameters.
+
+    :param adata: An AnnData object
+    :type adata: AnnData
+    :param channels: Channels/markers to normalize
+    :type channels: list
+    :param m: Logicle parameter. Default=4.5
+    :type m: float
+    :param q: Quantile to use for negative values. Default=0.05
+    :type q: float
+    """
     assert isinstance(adata, ad.AnnData), "Please provide an AnnData object"
     assert isinstance(channels, list), "Please provide a list of channels"
     channels = list(get_markers(adata, channels).keys())

src/FlowSOM/tl/_consensus_clustering.py

@@ -15,7 +15,9 @@
 
 class ConsensusCluster:
     """
-    Implementation of Consensus clustering, following the paper
+    Implementation of Consensus clustering.
+
+    This follows the paper
     https://link.springer.com/content/pdf/10.1023%2FA%3A1023949509487.pdf
     https://github.com/ZigaSajovic/Consensus_Clustering/blob/master/consensusClustering.py
       * cluster -> clustering class
@@ -36,7 +38,8 @@ def __init__(self, cluster, K, H, resample_proportion=0.9, linkage="average"):
         self.H_ = H
 
     def _internal_resample(self, data, proportion):
-        """
+        """Resamples the data.
+
         Args:
           * data -> (examples,attributes) format
           * proportion -> percentage to sample.

tests/test_basic.py

@@ -120,19 +120,19 @@ def test_FlowSOM_class(FlowSOM_res):
     var_true_keys_cell = ["pretty_colnames", "cols_used", "markers", "channels"]
 
     # make sure that at least the true keys are in the object
-    uns_keys_cell = all([key in cell_data.uns.keys() for key in uns_true_keys_cell])
-    obs_keys_cell = all([key in cell_data.obs.keys() for key in obs_true_keys_cell])
-    var_keys_cell = all([key in cell_data.var.keys() for key in var_true_keys_cell])
+    uns_keys_cell = all(key in cell_data.uns.keys() for key in uns_true_keys_cell)
+    obs_keys_cell = all(key in cell_data.obs.keys() for key in obs_true_keys_cell)
+    var_keys_cell = all(key in cell_data.var.keys() for key in var_true_keys_cell)
 
     cluster_data = FlowSOM_res.get_cluster_data()
     uns_true_keys_cluster = ["outliers", "graph", "xdim", "ydim", "metacluster_MFIs"]
     obs_true_keys_cluster = ["percentages", "metaclustering"]
     obsm_true_keys_cluster = ["cv_values", "sd_values", "mad_values", "codes", "grid", "layout"]
 
     # make sure that at least the true keys are in the object
-    uns_keys_cluster = all([key in cluster_data.uns.keys() for key in uns_true_keys_cluster])
-    obs_keys_cluster = all([key in cluster_data.obs.keys() for key in obs_true_keys_cluster])
-    obsm_keys_cluster = all([key in cluster_data.obsm.keys() for key in obsm_true_keys_cluster])
+    uns_keys_cluster = all(key in cluster_data.uns.keys() for key in uns_true_keys_cluster)
+    obs_keys_cluster = all(key in cluster_data.obs.keys() for key in obs_true_keys_cluster)
+    obsm_keys_cluster = all(key in cluster_data.obsm.keys() for key in obsm_true_keys_cluster)
     assert all(
         [
             uns_keys_cell,