def RunCellAssign(sce, annot_sce):
_rho_csv = os.path.join(os.path.split(sce)[0], "rho_csv_sub.csv")
_fit = os.path.join(os.path.split(sce)[0], "fit_sub.pkl")
sampleid = sce.split("/")[-2]
CellAssign.run(sce, config.rho_matrix, _fit, rho_csv=_rho_csv)
filtered_sce = os.path.join(os.path.split(sce)[0], "sce_cas.rdata")
shutil.copyfile(filtered_sce, annot_sce)
def test_cell_assign_em(self):
example_rda = os.path.join(base_dir, "tests/cell_assign_test.RData")
sce = SingleCellExperiment.fromRData(example_rda)
cellassigner = CellAssign()
rho = GeneMarkerMatrix(genes=[
"Gene161", "Gene447", "Gene519", "Gene609", "Gene677", "Gene750",
"Gene754", "Gene860", "Gene929", "Gene979"
],
cells=["Groups1", "Groups2"])
res = cellassigner.run_em(sce, rho)
def Run(sampleid, before, finished):
tenx = TenxDataStorage(sampleid, version="v3")
tenx.download()
analysis_path = tenx.tenx_path
tenx_analysis = TenxAnalysis(analysis_path)
tenx_analysis.load()
tenx_analysis.extract()
qc = QualityControl(tenx_analysis, sampleid)
CellAssign.run(qc.sce, config.rho_matrix,
".cache/{}/celltypes.rdata".format(sampleid))
open(finished, "w").write("Completed")
def test_cell_assign_pkl(self):
import pickle
import collections
tenx = TenxAnalysis("tests/pre_igo")
sce = TenX.read10xCounts(tenx)
handle = open("tests/rho_up.pkl","rb")
rho_matrix = pickle.load(handle)
handle.close()
rho = GeneMarkerMatrix(rho_matrix)
cellassigner = CellAssign()
res = cellassigner.run_em(sce, rho)
def test_cell_assign_em(self):
#example_rda = os.path.join(base_dir, "tests/cell_assign_test.RData")
print("Init CellEM")
sce = "sce_final.rdata"
# sce = SingleCellExperiment.fromRData(example_rda)
cellassigner = CellAssign()
# rho_matrix = dict()
# rho_matrix["Group1"] = ["Gene161", "Gene447", "Gene609", "Gene754", "Gene860", "Gene929", "Gene979"]
# rho_matrix["Group2"] = ["Gene161", "Gene447", "Gene609", "Gene754", "Gene860", "Gene929", "Gene979","Gene101","Gene212","Gene400"]
# rho = GeneMarkerMatrix(rho_matrix)
res = cellassigner.run_em(sce, "cell_assign_fit.rdata", "test")
def RunCellAssign(custom_output, sce, annot_sce):
sample = json.loads(open(custom_output,"r").read())
sampleid, path = list(sample.items()).pop()
temp = os.path.split(annot_sce)[0]
_rho_csv = os.path.join(os.path.split(sce)[0],"rho_csv_sub.csv")
_fit = os.path.join(os.path.split(sce)[0],"fit_sub.pkl")
_filtered_sce = os.path.join(os.path.split(sce)[0],"sce_cas.rdata")
filtered_sce = os.path.join(config.jobpath,"results","sce_cas_{}.rdata".format(sampleid))
if not os.path.exists(filtered_sce):
rho = os.path.join(config.jobpath,config.rho_matrix)
CellAssign.run(sce, rho, _fit, rho_csv=_rho_csv,lsf=True)
shutil.copyfile(_filtered_sce, filtered_sce)
shutil.copyfile(filtered_sce, annot_sce)
def RunHRD(custom_input, sce, rdata, umap):
sample = json.loads(open(custom_input, "r").read())
sampleid, path = list(sample.items()).pop()
rdata_cached = os.path.join(config.jobpath, "results",
"hrd_{}.rdata".format(sampleid))
umap_cached = os.path.join(config.jobpath, "results",
"hrd_umap_{}.png".format(sampleid))
temp = os.path.split(sce)[0]
rho_csv = os.path.join(os.path.split(sce)[0], "hrd_rho_csv.csv")
fit = os.path.join(os.path.split(sce)[0], "fit_sub_hrd.pkl")
filtered_sce = os.path.join(os.path.split(sce)[0], "hrd_sce_cas.rdata")
if not os.path.exists(rdata_cached):
rho = "/codebase/hrd.yaml"
CellAssign.run(sce,
rho,
fit,
rho_csv=rho_csv,
lsf=True,
script_prefix="hrd_")
shutil.copyfile(filtered_sce, rdata_cached)
rcode = """
library(Seurat)
library(ggplot2)
fit <- readRDS("{fit}")
sce <- readRDS("{sce}")
sce$repairtype <- fit$cell_type
sce$HRD_prob <- fit$mle_params$gamma[,"HRD"]
sce$HR_prob <- fit$mle_params$gamma[,"HR"]
seurat <- as.Seurat(sce, counts = "counts", data = "logcounts")
seurat$repairtype <- sce$repairtype
png("{umap}", width=1000, height=1000)
DimPlot(object = seurat, reduction = "UMAP", group.by = "repairtype", dark.theme=TRUE, plot.title="{sample}")
dev.off()
saveRDS(sce, file="{sce}")
"""
qc_script = os.path.join(temp, "hrd_viz_{}.R".format(sampleid))
output = open(qc_script, "w")
output.write(
rcode.format(sce=rdata_cached,
umap=umap_cached,
sample=sampleid,
fit=fit))
output.close()
subprocess.call(["Rscript", "{}".format(qc_script)])
shutil.copyfile(rdata_cached, rdata)
shutil.copyfile(umap_cached, umap)
def RunCellAssign(sce, annot_sce):
filtered_sce = os.path.join(os.path.split(sce)[0], "sce_cas.rdata")
_rho_csv = os.path.join(os.path.split(sce)[0], "rho_csv_sub.csv")
_fit = os.path.join(os.path.split(sce)[0], "fit.rdata")
sampleid = sce.split("/")[-2]
if not os.path.exists(filtered_sce):
CellAssign.run(sce,
config.rho_matrix,
_fit,
rho_csv=_rho_csv,
lsf=False)
shutil.copyfile(filtered_sce, annot_sce)
path = os.getcwd()
shutil.copyfile(_fit, os.path.join(path, "fit.rdata"))
def RunCellAssign(sce, annot_sce, cellfit):
_rho_csv = os.path.join(os.path.split(sce)[0], "rho_csv_sub.csv")
_fit = os.path.join(os.path.split(sce)[0], "fit_sub.pkl")
sampleid = sce.split("/")[-2]
filtered_sce = os.path.join(os.path.split(sce)[0], "sce_cas.rdata")
if not os.path.exists(filtered_sce) or not os.path.exists(_fit):
if "CD45N" in sampleid:
rho = config.negative_rho_matrix
elif "CD45P" in sampleid:
rho = config.positive_rho_matrix
else:
rho = config.rho_matrix
CellAssign.run(sce, rho, _fit, rho_csv=_rho_csv)
shutil.copyfile(filtered_sce, annot_sce)
shutil.copyfile(_fit.replace("fit_sub", "cell_types"), cellfit)
def RunExhaustion(custom_input, sce, rdata, umap):
sample = json.loads(open(custom_input, "r").read())
sampleid, path = list(sample.items()).pop()
rdata_cached = os.path.join(config.jobpath, "results",
"exhaustion_{}.rdata".format(sampleid))
umap_cached = os.path.join(config.jobpath, "results",
"exhaustion_umap_{}.png".format(sampleid))
temp = os.path.split(sce)[0]
_rho_csv = os.path.join(os.path.split(sce)[0], "exhaustion_rho_csv.csv")
_fit = os.path.join(os.path.split(sce)[0], "fit_sub_exhaustion.pkl")
_filtered_sce = os.path.join(
os.path.split(sce)[0], "exhaustion_sce_cas.rdata")
if not os.path.exists(rdata_cached):
rho = "/codebase/exhaustion.yaml"
CellAssign.run(sce,
rho,
_fit,
rho_csv=_rho_csv,
lsf=True,
script_prefix="exhaustion_")
shutil.copyfile(_filtered_sce, rdata_cached)
shutil.copyfile(rdata_cached, rdata)
rcode = """
library(Seurat)
library(ggplot2)
sce <- readRDS("{sce}")
fit <- readRDS("{fit}")
sce$Exhaustion_prob <- fit$mle_params$gamma[,"Exhausted.T.cell"]
seurat <- as.Seurat(sce, counts = "counts", data = "logcounts")
seurat$Exhaustion_prob <- sce$Exhaustion_prob
png("{umap}", width=1000, height=1000)
FeaturePlot(object = seurat, reduction = "UMAP", features=c("Exhaustion_prob"))
dev.off()
"""
qc_script = os.path.join(temp, "exhaustion_viz_{}.R".format(sampleid))
output = open(qc_script, "w")
output.write(
rcode.format(sce=rdata, umap=umap_cached, sample=sampleid, fit=_fit))
output.close()
subprocess.call(["Rscript", "{}".format(qc_script)])
shutil.copyfile(umap_cached, umap)
def Run(sampleid, before, finished, use_corrected=False):
if use_corrected and os.path.exists(".cache/corrected/"):
sce = ".cache/corrected/corrected_sce.rdata"
if not os.path.exists(sce):
utils = DropletUtils()
utils.read10xCounts(".cache/corrected/",
".cache/corrected/corrected_sce.rdata")
else:
tenx = TenxDataStorage(sampleid, version="v3")
tenx.download()
analysis_path = tenx.tenx_path
tenx_analysis = TenxAnalysis(analysis_path)
tenx_analysis.load()
tenx_analysis.extract()
qc = QualityControl(tenx_analysis, sampleid)
sce = qc.sce
if not os.path.exists(".cache/{}/celltypes.rdata".format(sampleid)):
CellAssign.run(sce, config.rho_matrix,
".cache/{}/celltypes.rdata".format(sampleid))
open(finished, "w").write("Completed")
def Run(sampleid, raw_sce, sce_cas, rho, B, min_delta):
output = os.path.join(os.path.split(raw_sce)[0], "sce_cas.rdata")
celltypes = os.path.join(os.path.split(raw_sce)[0], "celltypes.rdata")
if not os.path.exists(celltypes):
CellAssign.run(raw_sce, rho, celltypes, B, min_delta)
shutil.copyfile(output, sce_cas)