from progress.bar import Bar
import numpy as np
px.print_libraries()
libs = px.list_libraries()
gmt_file = px.load_library(libs[28])
outname = libs[28]
correlationFolder = "correlation_100_folder"
predictionFolder = "prediction_100_folder"
outfolder = "prismxresult_100"
px.predict_gmt("gobp_model_100.pkl",
gmt_file,
correlationFolder,
predictionFolder,
outfolder,
outname,
step_size=1000,
verbose=True)
geneAUC, setAUC = px.benchmarkGMT(gmt_file,
correlationFolder,
predictionFolder,
outfolder + "/" + outname + ".f",
verbose=True)
fig, ax = plt.subplots(figsize=[4, 16])
violin_parts = ax.violinplot(setAUC.values,
range(0, setAUC.shape[1]),
points=200,
vert=False,
verbose=True)
pickle.dump(model, open(lib + "_model_" + str(clusterCount) + ".pkl",
'wb'))
outfolder = "prismxresult_" + str(clusterCount)
os.makedirs("test_data", exist_ok=True)
for lib in genesetlibs[0:2]:
for lib2 in genesetlibs[0:6]:
outname = lib2
gmt_file = px.load_library(lib2)
px.predict_gmt(lib + "_model_" + str(clusterCount) + ".pkl",
gmt_file,
correlationFolder,
predictionFolder,
outfolder,
outname,
step_size=200,
intersect=True,
verbose=True)
# benchmark the prediction quality
geneAUC, setAUC = px.benchmarkGMTfast(gmt_file,
correlationFolder,
predictionFolder,
outfolder + "/" + outname + ".f",
intersect=True,
verbose=True)
geneAUC = geneAUC.reset_index()
setAUC = setAUC.reset_index()
geneAUC.to_feather("test_data/auc_gene_" + lib + "_" + lib2 + ".f")
setAUC.to_feather("test_data/auc_set_" + lib + "_" + lib2 + ".f")