def annotate_doublets(mtx_fpath, feature_fpath, expected_doublet_rate2=0.06):
if False:
plt.rcParams['font.family'] = 'sans-serif'
plt.rcParams['font.sans-serif'] = 'Arial'
plt.rc('font', size=14)
plt.rcParams['pdf.fonttype'] = 42
counts_matrix = scipy.io.mmread(mtx_fpath).T.tocsc()
genes = np.array(scr.load_genes(feature_fpath, delimiter='\t', column=1))
print('Counts matrix shape: {} rows, {} columns'.format(
counts_matrix.shape[0], counts_matrix.shape[1]))
print('Number of genes in gene list: {}'.format(len(genes)))
scrub = scr.Scrublet(counts_matrix,
expected_doublet_rate=expected_doublet_rate2)
doublet_scores, predicted_doublets = scrub.scrub_doublets(
min_counts=2,
min_cells=3,
min_gene_variability_pctl=85,
n_prin_comps=30)
if False:
scrub.plot_histogram()
print('Running UMAP...')
scrub.set_embedding(
'UMAP', scr.get_umap(scrub.manifold_obs_, 10, min_dist=0.3))
print('Done.')
scrub.plot_embedding('UMAP', order_points=True)
return ([doublet_scores, predicted_doublets])
def set_embedding(self):
print("#### Run UMAP")
self.scrub.set_embedding(
'UMAP', scr.get_umap(self.scrub.manifold_obs_, 30, min_dist=0.3))
# # Uncomment to run tSNE - slow
print('#### Running tSNE...')
self.scrub.set_embedding(
'tSNE',
scr.get_tsne(self.scrub.manifold_obs_, angle=0.9, verbose=True))
print('Done.')
with open(os.path.join(self._outdir, "scrub.pkl"), "wb") as f:
pickle.dump(self.scrub, f)
f.close()
def scrublet_c(sample, inDir, outDir, expected_doublet_rate, sim_doublet_ratio,
ratio_df, out_df):
print(sample, "start scrublet")
counts_matrix = scipy.io.mmread(os.path.join(inDir,
'matrix.mtx')).T.tocsc()
genes = np.array(
scr.load_genes(os.path.join(inDir, 'genes.tsv'),
delimiter='\t',
column=1))
scrub = scr.Scrublet(counts_matrix,
expected_doublet_rate=expected_doublet_rate,
sim_doublet_ratio=sim_doublet_ratio)
doublet_scores, predicted_doublets = scrub.scrub_doublets(
min_counts=2,
min_cells=3,
min_gene_variability_pctl=85,
n_prin_comps=30)
scrub.plot_histogram()
plt.savefig(
os.path.join(
outDir, "{0}_scrublet_doublet_score_histogram.pdf".format(sample)))
print(sample, 'Running scrublet UMAP...')
scrub.set_embedding('UMAP',
scr.get_umap(scrub.manifold_obs_, 10, min_dist=0.3))
print(sample, 'scrublet Done.')
scrub.plot_embedding('UMAP', order_points=True)
plt.savefig(os.path.join(outDir, "{0}_scrublet_UMAP.pdf".format(sample)))
print(sample, "Done scrublet")
ratio_df.loc['scrublet', sample] = scrub.detected_doublet_rate_
out_df['scrublet_doublet_scores'] = doublet_scores
out_df['scrublet_doublets'] = predicted_doublets
return ratio_df, out_df
#write scrublet output to file:
print("Writing doublet predictions to %s" % args.output + "/" + args.name + "_predicted_doublets.tsv", file=sys.stderr)
with open(args.output + "/" + args.name + "_predicted_doublets.tsv", 'w') as outfile:
outfile.write("\t".join(["barcode", "doublet_score", "doublet_prediction"])+"\n")
for barcode, score, prediction in zip(barcodes, doublet_scores, predicted_doublets):
if prediction == False:
doublet = "0"
else:
doublet = "1"
outfile.write("\t".join([barcode, str(score), doublet])+"\n")
print("Plotting doublet score histogram to %s" % args.output + "/" + args.name + "_score_histogram.pdf", file=sys.stderr)
f = scrub.plot_histogram()
plt.savefig(args.output + "/" + args.name + "_score_histogram.pdf")
if args.embed == True:
print("Running UMAP", file=sys.stderr)
scrub.set_embedding('UMAP', scr.get_umap(scrub.manifold_obs_, 10, min_dist=0.3))
print("Plotting UMAP to %s" % args.output + "/" + args.name + "_UMAP.pdf", file=sys.stderr)
f = scrub.plot_embedding('UMAP', order_points=True);
plt.savefig(args.output + "/" + args.name + "_UMAP.pdf")
print("Running TSNE", file=sys.stderr)
scrub.set_embedding('tSNE', scr.get_tsne(scrub.manifold_obs_, angle=0.9))
print("Plotting TSNE to %s" % args.output + "/" + args.name + "_TSNE.pdf", file=sys.stderr)
f = scrub.plot_embedding('tSNE', order_points=True);
plt.savefig(args.output + "/" + args.name + "_TSNE.pdf")
predicted_doublets[sample] = scrub[sample].call_doublets(threshold=0.25)
sum(predicted_doublets[sample])
sample = 'SIGAG2'
scrub[sample].plot_histogram()
predicted_doublets[sample] = scrub[sample].call_doublets(threshold=0.35)
sum(predicted_doublets[sample])
for sample in samples:
print(sample, ':', sum(predicted_doublets[sample]))
# Iteratively adjust thresholds according to plot results
sample = 'SIGAE2'
scrub[sample].set_embedding(
'UMAP', scr.get_umap(scrub[sample].manifold_obs_, 10, min_dist=0.3))
scrub[sample].plot_embedding('UMAP', order_points=True)
#scrub[sample].predicted_doublets_ = scrub[sample].call_doublets(threshold=0.3)
#scrub[sample].call_doublets(threshold=0.3).sum()
#predicted_doublets[sample] = scrub[sample].call_doublets(threshold=0.3)
sample = 'SIGAF2'
scrub[sample].set_embedding(
'UMAP', scr.get_umap(scrub[sample].manifold_obs_, 10, min_dist=0.25))
scrub[sample].plot_embedding('UMAP', order_points=True)
#scrub[sample].predicted_doublets_ = scrub[sample].call_doublets(threshold=0.25)
#scrub[sample].call_doublets(threshold=0.25).sum()
#predicted_doublets[sample] = scrub[sample].call_doublets(threshold=0.25)
sample = 'SIGAG2'
scrub[sample].set_embedding(
out_df = pd.DataFrame({
'Cell.ID': cells['Cell.ID'],
'scrublet_doublet_score': doublet_scores,
'scrublet_doublet_call1': predicted_doublets,
'scrublet_doublet_call2': predicted_doublets_025
})
out_df.to_csv('%s.scrublet_out.csv' % (prefix), sep="\t", index=False)
with PdfPages('%s.scrublet_out.pdf' % (prefix)) as pdf:
scrub.plot_histogram()
pdf.savefig()
print('Running UMAP...')
scrub.set_embedding('UMAP',
scr.get_umap(scrub.manifold_obs_, 10, min_dist=0.3))
# # Uncomment to run tSNE - slow
# print('Running tSNE...')
# scrub.set_embedding('tSNE', scr.get_tsne(scrub.manifold_obs_, angle=0.9))
# # Uncomment to run force layout - slow
# print('Running ForceAtlas2...')
# scrub.set_embedding('FA', scr.get_force_layout(scrub.manifold_obs_, n_neighbors=5. n_iter=1000))
pdf.savefig()
print('Done.')
scrub.plot_embedding('UMAP', order_points=True)
# scrub.plot_embedding('tSNE', order_points=True);
# scrub.plot_embedding('FA', order_points=True);
pdf.savefig()
def main():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.RawDescriptionHelpFormatter,
epilog='author: {0} mail: {1}'.format(__author__, __mail__))
parser.add_argument('-m',
'--mtx',
help='cellranger分析结果中的matrix.mtx',
dest='mtx',
required=True)
parser.add_argument('-f',
'--feature',
help='cellranger分析结果中的feature.csv或genes.csv',
dest='feature',
required=True)
parser.add_argument('-o',
'--outdir',
help='结果输出目录',
dest='outdir',
required=True)
parser.add_argument('-s',
'--sampleName',
help='样本名',
dest='sampleName',
required=True)
parser.add_argument('-e',
'--expectedDoubletRate',
help='细胞结团率',
dest='expectedDoubletRate',
type=float,
required=True)
parser.add_argument('-p',
'--pc',
help='PC值',
dest='pc',
type=int,
default=30)
args = parser.parse_args()
logging.basicConfig(
level=logging.DEBUG,
format=
"%(asctime)s - %(filename)s[line:%(lineno)d] - %(levelname)s - %(message)s"
)
logging.info("开始分析")
# plt.rcParams['font.family'] = 'sans-serif'
# plt.rcParams['font.sans-serif'] = 'Arial'
# plt.rc('font', size=14)
# plt.rcParams['pdf.fonttype'] = 42
pc = args.pc
expectedDoubletRate = args.expectedDoubletRate
sampleName = args.sampleName
#Load counts matrix and gene list
counts_matrix = scipy.io.mmread(args.mtx).T.tocsc()
genes = np.array(scr.load_genes(args.feature, delimiter='\t', column=1))
scrub = scr.Scrublet(counts_matrix,
expected_doublet_rate=expectedDoubletRate)
#Run the default pipeline
doublet_scores, predicted_doublets = scrub.scrub_doublets(
min_counts=2,
min_cells=3,
min_gene_variability_pctl=85,
n_prin_comps=pc)
#Plot doublet score histograms for observed transcriptomes and simulated doublets
scrub.call_doublets(threshold=0.1)
scrub.plot_histogram()
plt.savefig(args.outdir + "/" + sampleName + '_Scrublet_Histogram.png')
scrub.set_embedding('UMAP',
scr.get_umap(scrub.manifold_obs_, 10, min_dist=0.3))
scrub.plot_embedding('UMAP', order_points=True)
plt.savefig(args.outdir + "/" + sampleName + '_Scrublet_UMAP.png')
#output the log file
#expected doublet rate、 detected doublet rate、 doublet threshold、overall doublet rate
logging.info(
"patientID\texpected_doublet_rate\tdetected_doublet_rate\toverall_doublet_rate\tthreshold\tPC\n"
)
logging.info(
"%s\t%.4f\t%.4f\t%.4f\t%.4f\t%s\n" %
(sampleName, scrub.expected_doublet_rate, scrub.detected_doublet_rate_,
scrub.overall_doublet_rate_, scrub.threshold_, pc))
#output the doublet status of every single cell
with open(args.outdir + "/" + sampleName + ".predictDoublet_scrublet.txt",
"w+") as fo:
for i in scrub.predicted_doublets_:
fo.write("%s\n" % (i))
counts_matrix = scipy.io.mmread(inputfolder + '/' + jobname +
'_filtered.mtx').T.tocsc()
print('Counts matrix shape: {} rows, {} columns'.format(
counts_matrix.shape[0], counts_matrix.shape[1]))
scrub = scr.Scrublet(counts_matrix, expected_doublet_rate=exp_db_rate)
doublet_scores, predicted_doublets = scrub.scrub_doublets(
min_counts=2, min_cells=3, min_gene_variability_pctl=85, n_prin_comps=npcs)
scrub.call_doublets(threshold=db_thr)
# Duplet score for cells
savetxt(outputfolder + '/' + jobname + '_duplets_score.csv',
scrub.doublet_scores_obs_,
delimiter=',')
# Simulated duplets
savetxt(outputfolder + '/' + jobname + '_sim_duplets_score.csv',
scrub.doublet_scores_sim_,
delimiter=',')
# UMAP
print('Running UMAP...')
umap = scr.get_umap(scrub.manifold_obs_, n_neighbors=10, min_dist=0.1)
savetxt(outputfolder + '/' + jobname + '_umap_scrublet.csv',
umap,
delimiter=',')
print('Done.')