def rebuild_comparisons(allsorts_clf, probabilities, ui, size=10):
X_filtered = allsorts_clf.transform(ui.samples)["counts"]
''' Probability comparisons '''
true_samples = []
for sample, column in probabilities.iterrows():
if column["True"] in column["Pred"]:
true_samples.append(sample)
comparisons = probabilities.loc[true_samples]
cgroup = comparisons.groupby("True")
new_samples = cgroup.apply(lambda x: x.sample(n=size) if x.shape[0] > size
else x.sample(n=x.shape[0]))
new_index = []
for sample in new_samples.index:
new_index.append(sample[1])
filename = "comparisons.csv"
destination = str(root_dir()) + "/models/allsorts/" + filename
cfinal = comparisons[(comparisons.index.isin(new_index))]
cfinal.to_csv(destination)
'''Umap Visualisation'''
# Get genes used in the final models and filter counts by them
chosen_genes = []
classifiers = allsorts_clf.steps[-1][-1].fitted
feature_selection = allsorts_clf.steps[-3][-1].genes
for subtype, clf in classifiers.items():
genes = feature_selection[subtype]
sub_genes = pd.DataFrame(clf.coef_, columns=genes)
sub_genes = sub_genes[sub_genes != 0].dropna(axis=1)
chosen_genes += list(sub_genes.columns)
chosen_genes = list(set(chosen_genes))
X_filtered = X_filtered.loc[true_samples, chosen_genes]
labels = ui.labels.loc[true_samples]
# Create UMAPs
u = UMAP(n_neighbors=10).fit(X_filtered)
# Save
u_filename = 'umap.sav'
l_filename = 'comparison_labels.csv'
g_filename = 'comparison_genes.csv'
destination = str(
root_dir()) + "/models/allsorts/comparisons/" + u_filename
joblib.dump(u, destination)
labels.to_csv(
str(root_dir()) + "/models/allsorts/comparisons/" + l_filename)
pd.Series(X_filtered.columns).to_csv(
str(root_dir()) + "/models/allsorts/comparisons/" + g_filename)
def _loadChrom(self):
chrom_ref_path = str(root_dir())+"/data/chrom_refs.txt"
self.chrom_ref = pd.read_csv(chrom_ref_path, sep="\t", header=None)
self.chrom_ref.drop([1, 2, 5, 6, 7], axis=1, inplace=True)
self.chrom_ref.columns = ["chrom", "start", "end", "meta"]
self.chrom_ref["length"] = (self.chrom_ref["start"] -
self.chrom_ref["end"]).abs()
# Extract gene names
gene_names = []
for line, meta in self.chrom_ref["meta"].iteritems():
gene_name = meta.split(";")[2].split('"')[1]
gene_names.append(gene_name)
self.chrom_ref["gene"] = gene_names
self.chrom_ref.index = self.chrom_ref["chrom"]
self.chrom_ref.drop(["chrom", "meta"], axis=1, inplace=True)
self.chrom_ref["start"] = pd.to_numeric(self.chrom_ref["start"])
self.chrom_ref["end"] = pd.to_numeric(self.chrom_ref["end"])
# Create dictionary of genes per chromosome
self.chrom_dict = {}
for chrom, info in self.chrom_ref.iterrows():
if chrom in self.chrom_dict:
self.chrom_dict[chrom].append(info["gene"])
else:
self.chrom_dict[chrom] = [info["gene"]]
self.chroms = list(range(1, 23)) + ["X", "Y"]
def load_classifier(path=False):
"""
Load the ALLSorts classifier from a pickled file.
...
Parameters
__________
path : str
Path to a pickle object that holds the ALLSorts model.
Default: "/models/allsorts/allsorts.pkl.gz"
Returns
__________
allsorts_clf : ALLSorts object
ALLSorts object, unpacked, ready to go.
"""
if not path:
path = str(root_dir()) + "/models/allsorts/allsorts.pkl.gz"
message("Loading classifier...")
allsorts_clf = joblib.load(path)
return allsorts_clf
def __init__(self):
if self._is_cli():
self.cli = True
self.input = self._get_args()
self.samples = self.input.samples
self.labels = self.input.labels if self.input.labels else False
self.model_dir = str(
root_dir()
) + "/models/allsorts/" if not self.input.model_dir else self.input.model_dir
self.destination = False if not self.input.destination else self.input.destination
self.test = self.input.test
self.train = False if not self.input.train else True
self.comparison = False if not self.input.comparison else True
self.n_jobs = 1 if not self.input.njobs else int(self.input.njobs)
self.verbose = False if not self.input.verbose else True
self.force = False if not self.input.force else True
self.cv = 3 if not self.input.cv else int(self.input.cv)
self.parents = False if not self.input.parents else True
self.ball = self.input.ball
self._input_checks()
self._load_samples()
else:
message(
"No arguments supplied. Please use allsorts --help for further information about input."
)
sys.exit(0)
def get_figures(samples,
allsorts,
destination,
probabilities,
plots=["distributions", "waterfalls"]):
"""
Make figures of the results.
...
Parameters
__________
samples : Pandas DataFrame
Pandas DataFrame that represents the raw counts of your samples (rows) x genes (columns)).
destination : str
Location of where the results should be saved.
probabilities : Pandas DataFrame
The result of running the get_predictions(samples, labels=False, parents=False) function.
See function for further usage.
plots : List
List of plots required. Default: "distributions", "waterfalls", and "manifold".
See https://github.com/Oshlack/AllSorts/ for examples.
Output
__________
Distributions.png, Waterfalls.png, Manifold.png at the ui.destination path.
"""
message("Saving figures...")
for plot in plots:
if plot == "distributions":
dist_plot = allsorts.predict_dist(probabilities, return_plot=True)
dist_plot.savefig(destination + "/distributions.png")
if plot == "waterfalls":
if "True" in probabilities.columns:
comparisons = False
else:
comparisons = pd.read_csv(str(root_dir()) +
"/models/allsorts/comparisons.csv",
index_col=0)
waterfall_plot = allsorts.predict_waterfall(probabilities,
compare=comparisons,
return_plot=True)
waterfall_plot.savefig(destination + "/waterfalls.png")
if plot == "manifold":
umap_plot = allsorts.predict_plot(samples, return_plot=True)
umap_plot.savefig(destination + "/manifold.png")
def predict_plot(self, X, return_plot=False):
"""
Given the raw counts, embed these within a UMAP visualisation consisting of the comparison data.
...
Parameters
__________
X : Pandas DataFrame
Pandas DataFrame that represents the raw counts of your samples (rows) x genes (columns)).
return_plot : bool
Rather than showing the plot through whatever IDE is being used, send it back to the function call.
Likely so it can be saved.
Returns
__________
Matplotlib object containing the drawn figure
Output
__________
UMAP Plot figure.
"""
plt.figure(figsize=(20, 10))
u = joblib.load(
str(root_dir()) + "/models/allsorts/comparisons/umap.sav")
c_labels = pd.read_csv(
str(root_dir()) +
"/models/allsorts/comparisons/comparison_labels.csv",
index_col=0)
c_labels = c_labels["labels"]
c_genes = pd.read_csv(
str(root_dir()) +
"/models/allsorts/comparisons/comparison_genes.csv",
index_col=0)
c_genes = list(c_genes.iloc[:, 0])
u_c = u.embedding_
X_t = self.transform(X)
X_t = X_t["counts"].loc[:, c_genes]
u_t = u.transform(X_t)
plt.scatter(u_t[:, 0], u_t[:, 1], c="#000000", alpha=0.4)
plt.scatter(u_c[:, 0],
u_c[:, 1],
c=[c_subtypes[r] for r in c_labels],
alpha=0.4,
marker="x")
transformed_positions = pd.DataFrame(u_c)
transformed_positions["label"] = list(c_labels)
median = transformed_positions.groupby("label").median()
for name, label in median.iterrows():
plt.text(label[0], label[1], name, FontSize=16)
if return_plot:
return plt
else:
plt.show()