def print_my_plots(r, rplots, out='', file_type='pdf'):
"""
saves our plots to files named with the plotting method used.
:param r: pointer to the R instance
:param rplots: the ``Bunch`` object where we stored our plots
:param out: a base directory to add to our saved plots into
:param file_type: the type of output file to use, choices: ['pdf','jpeg','png','ps']
"""
out = out.rstrip('/')
mkdirp(out)
for plot_id in rplots:
file_path = "%s/%s.%s" % (out,plot_id,file_type)
r.ggsave(filename=file_path,plot=rplots[plot_id])
def KM(OS, Censored, as_group, data, ggsave=False, path="./", pvalue=0):
# 分组
surv_data = data[[OS, Censored, as_group]].sort_values(by=[as_group])
surv_data["group"] = "L"
surv_data.iloc[int(surv_data.shape[0] / 2):]["group"] = "H"
# Kaplan-Meier生存曲线
with localconverter(ro.default_converter + pandas2ri.converter):
robjects.globalenv["surv_data"] = pandas2ri.py2rpy(surv_data)
robjects.globalenv["surv_diff"] = r(
f"survdiff(Surv({OS}, {Censored})~group,surv_data,rho = 0)")
Pvalue = r("1 - pchisq(surv_diff$chisq, length(surv_diff$n) -1)")[0]
if ggsave and Pvalue < pvalue:
r.ggsave(r(
f"autoplot(survfit(Surv({OS}, {Censored})~group,surv_data), xlab = 'Time', ylab = 'Survival')+ggtitle('Pvalue = {Pvalue}')"
),
file=f"{path}/{gene}.pdf")
return Pvalue
def collect(self, blocks, figure_key="", subfig=0):
'''collect one or more R figures.
Plots are collected from all active devices.
Plots are also collected from result-blocks
containing a 'ggplot' attribute.
1. save as png, hires-png and pdf
2. save thumbnail
3. insert rendering code at placeholders in output
returns a map of place holder to placeholder text.
'''
# disable plotting if no rpy installed
if R is None:
return {}
map_figure2text = {}
# determine the image formats to create
default_format, additional_formats = Utils.getImageFormats(
self.display_options)
all_formats = [default_format, ] + additional_formats
image_options = Utils.getImageOptions(self.display_options)
##########################################
##########################################
##########################################
# iterate over devices
devices = R["dev.list"]()
try:
maxid = max(R["dev.list"]())
except TypeError:
maxid = 0
for figid in range(2, maxid + 1):
for id, format, dpi in all_formats:
R["dev.set"](figid)
outname = "%s_%02d" % (self.template_name, figid)
outpath = os.path.join(self.outdir, '%s.%s' % (outname, format))
if format.endswith("png"):
# for busy images there is a problem with figure margins
# simply increase dpi until it works.
R["dev.set"](figid)
width = height = 480 * dpi / 80
x = 0
while 1:
try:
R["dev.copy"](device=R.png,
filename=outpath,
res=dpi,
width=width,
height=height)
R["dev.off"]()
except rpy2.rinterface.RRuntimeError:
width *= 2
height *= 2
if x < 5:
continue
break
elif format.endswith("svg"):
R["dev.copy"](device=R.svg,
filename=outpath)
R["dev.off"]()
elif format.endswith("eps"):
R["dev.copy"](device=R.postscript,
paper='special',
width=6,
height=6,
file=outpath,
onefile=True)
R["dev.off"]()
elif format.endswith("pdf"):
R["dev.copy"](device=R.pdf,
paper='special',
width=6,
height=6,
file=outpath,
onefile=True)
R["dev.off"]()
else:
raise ValueError("format '%s' not supported" % format)
if not os.path.exists(outpath):
continue
# raise ValueError("rendering problem: image file was not be created: %s" % outpath)
if format == 'png':
thumbdir = os.path.join(self.outdir, 'thumbnails')
try:
os.makedirs(thumbdir)
except OSError:
pass
thumbfile = str('%s.png' % os.path.join(thumbdir, outname))
captionfile = str(
'%s.txt' % os.path.join(thumbdir, outname))
if not os.path.exists(thumbfile):
# thumbnail only available in matplotlib >= 0.98.4
try:
figthumb = image.thumbnail(
str(outpath), str(thumbfile), scale=0.3)
except AttributeError:
pass
outfile = open(captionfile, "w")
outfile.write("\n".join(self.content) + "\n")
outfile.close()
R["dev.off"](figid)
# create the text element
rst_output = Utils.buildRstWithImage(outname,
self.outdir,
self.rstdir,
self.builddir,
self.srcdir,
additional_formats,
self.tracker_id,
self.links,
self.display_options)
map_figure2text["#$rpl %i$#" % figid] = rst_output
figid = maxid
##########################################
##########################################
##########################################
# iterate over ggplot plots
for block in blocks:
if not hasattr(block, "rggplot"):
continue
pp = block.rggplot
figname = block.figname
outname = "%s_%s" % (self.template_name, figname)
for id, format, dpi in all_formats:
outpath = os.path.join(self.outdir, '%s.%s' % (outname, format))
try:
R.ggsave(outpath, plot=pp, dpi=dpi)
except rpy2.rinterface.RRuntimeError as msg:
raise
# width, height = 3 * dpi, 3 * dpi
# if format.endswith("png"):
# R.png(outpath,
# width = width,
# height = height)
# elif format.endswith("svg"):
# R.svg(outpath)
# elif format.endswith("eps"):
# R.postscript(outpath)
# elif format.endswith("pdf"):
# R.pdf(outpath)
# R.plot(pp)
# R["dev.off"]()
# create the text element
rst_output = Utils.buildRstWithImage(outname,
self.outdir,
self.rstdir,
self.builddir,
self.srcdir,
additional_formats,
self.tracker_id,
self.links,
self.display_options)
map_figure2text["#$ggplot %s$#" % figname] = rst_output
return map_figure2text
################# 生存分析 #################
# https://www.jianshu.com/p/4ad9ba730719
# r('''suppressMessages(library(survival))''')
importr("survival")
importr("ggfortify")
with localconverter(ro.default_converter + pandas2ri.converter):
# 构建生存对象
robjects.globalenv["cox_data"] = ro.conversion.py2rpy(
sample[["Stromal_score", "Stromal_Group", "OS.time",
"OS"]].dropna())
robjects.globalenv["surv"] = r(
"Surv(time=cox_data$OS.time, event=cox_data$OS)")
# Kaplan-Meier生存曲线
KM_Stromal_fit = r("survfit(surv~cox_data$Stromal_Group)")
r.ggsave(r.autoplot(KM_Stromal_fit), file="KM_Stromal_fit.pdf")
# Long-rank检验(对数秩和检验)
print(r("survdiff(surv~cox_data$Stromal_Group,rho = 0)"))
# 单因素Cox
print(r("coxph(surv~cox_data$Stromal_score)"))
################# 基因差异表达分析 #################
## 读取表达矩阵
with localconverter(ro.default_converter + pandas2ri.converter):
exprSet_R = ro.conversion.py2rpy(HT_HG_U133A_sample)
group_list = "-".join(
list(sample["Stromal_Group"].astype("str").unique())[::-1])
#exprSet_R = r(f'''read.table("{path}HT_HG_U133A_sample.txt",header=TRUE,sep="\t",row.names="sample")''')
## 制作分组矩阵
design = sample[["Stromal_Group_L",
"Stromal_Group_H"]].set_index(sample["ID"])
