def gsea(homepath):
'''
Parameters
----------
`homepath` (str):
Path where you want to save all the generated files
and folders.
Return:
-------
None
Outputs:
--------
Generate a directory names enrichr
within home directory and two plot
of gene enrichement analysis using
the selected genes from panclassif
'''
warnings.filterwarnings("ignore")
# Directory
directory = "enrichr"
# Parent Directory path
parent_dir = homepath
# Path
path = os.path.join(parent_dir, directory)
if not os.path.exists(path):
os.mkdir(path)
gene = pd.read_csv(homepath + "/std_npy/unique_genes_with_frequency.csv",
header=None)
gl = []
for g in range(len(gene)):
gl.append(gene[0][g])
enr = gs.enrichr(gene_list=gl,
description='Disease',
gene_sets='DisGeNET',
outdir=homepath + '/enrichr')
# simple plotting function
from gseapy.plot import barplot, dotplot
# to save your figure, make sure that ``ofname`` is not None
barplot(enr.res2d,
title='DisGeNET',
cutoff=0.2,
ofname=homepath + '/enrichr/DisGeNET_barplot.png')
dotplot(enr.res2d,
title='DisGeNET',
cmap='viridis_r',
cutoff=0.2,
ofname=homepath + '/enrichr/DisGeNET_dotplot.png')
def run(self):
"""run enrichr for one sample gene list but multi-libraries"""
# read input file
genes_list = self.parse_genelists()
gss = self.parse_genesets()
# if gmt
self._logger.info(
"Connecting to Enrichr Server to get latest library names")
if len(gss) < 1:
self._logger.error("Hint: Current organism = %s, is this correct?\n"%self.organism +\
"Hint: use get_library_name() to view full list of supported names.")
raise LookupError(
"Not validated Enrichr library! Please provide correct organism and library name!"
)
self.results = pd.DataFrame()
for g in gss:
if isinstance(g, dict):
## local mode
res = self.enrich(g)
shortID, self._gs = str(id(g)), "CUSTOM%s" % id(g)
if res is None:
self._logger.info(
"No hits return, for gene set: Custom%s" % shortID)
continue
else:
## online mode
self._gs = str(g)
self._logger.debug("Start Enrichr using library: %s" %
(self._gs))
self._logger.info('Analysis name: %s, Enrichr Library: %s' %
(self.descriptions, self._gs))
shortID, res = self.get_results(genes_list)
# Remember gene set library used
res.insert(0, "Gene_set", self._gs)
# Append to master dataframe
self.results = self.results.append(res, ignore_index=True)
self.res2d = res
if self._outdir is None: continue
self._logger.info('Save file of enrichment results: Job Id:' +
str(shortID))
outfile = "%s/%s.%s.%s.reports.txt" % (self.outdir, self._gs,
self.organism, self.module)
self.res2d.to_csv(outfile, index=False, encoding='utf-8', sep="\t")
# plotting
if not self.__no_plot:
msg = barplot(df=res,
cutoff=self.cutoff,
figsize=self.figsize,
top_term=self.__top_term,
color='salmon',
title=self._gs,
ofname=outfile.replace("txt", self.format))
if msg is not None: self._logger.warning(msg)
self._logger.info('Done.\n')
# clean up tmpdir
if self._outdir is None: self._tmpdir.cleanup()
return
def run(self):
"""run enrichr for one sample gene list but multi-libraries"""
# read input file
genes_list = self.parse_genelists()
gss = unique(self.parse_genesets())
self._logger.info(
"Connecting to Enrichr Server to get latest library names")
# gss = self.gene_sets.split(",")
enrichr_library = get_library_name()
gss = [g for g in gss if g in enrichr_library]
self._logger.info("Libraries are used: %s" % ("',".join(gss)))
if len(gss) < 1:
sys.stderr.write("Not validated Enrichr library name provided\n")
sys.stdout.write(
"Hint: use get_library_name() to view full list of supported names"
)
sys.exit(1)
self.results = pd.DataFrame()
for g in gss:
self._gs = str(g)
self._logger.debug("Start Enrichr using library: %s" % (self._gs))
self._logger.info('Analysis name: %s, Enrichr Library: %s' %
(self.descriptions, self._gs))
shortID, res = self.get_results(genes_list)
# Remember gene set library used
res.insert(0, "Gene_set", self._gs)
# Append to master dataframe
self.results = self.results.append(res, ignore_index=True)
self.res2d = res
if self._outdir is None: continue
self._logger.info('Save file of enrichment results: Job Id:' +
str(shortID))
outfile = "%s/%s.%s.%s.reports.txt" % (
self.outdir, self._gs, self.descriptions, self.module)
self.res2d.to_csv(outfile, index=False, encoding='utf-8', sep="\t")
# plotting
if not self.__no_plot:
msg = barplot(df=res,
cutoff=self.cutoff,
figsize=self.figsize,
top_term=self.__top_term,
color='salmon',
title=self._gs,
ofname=outfile.replace("txt", self.format))
if msg is not None: self._logger.warning(msg)
self._logger.info('Done.\n')
# clean up tmpdir
if self._outdir is None: self._tmpdir.cleanup()
return
import sleep
import matplotlib.pyplot as plt
from gseapy.parser import Biomart
from gseapy.plot import barplot, dotplot
gene_list = pd.read_csv("/Users/sunxueyan/Downloads/GSEApy-master/tests/data/gene_list.txt",header=None, sep="\t")
gene_list1 = pd.read_csv("/Users/sunxueyan/Downloads/non_geneID.csv")
gene_list1.head()
glist = gene_list1.squeeze().str.strip().tolist()
names = gp.get_library_name() # default: Human
s = requests.session()
s.keep_alive = False
enr = gp.enrichr(gene_list="/Users/sunxueyan/Downloads/GSEApy-master/tests/data/gene_list.txt",
# or gene_list=glist
description='',
gene_sets=['KEGG_2019_Human'],
outdir='test/enrichr_kegg',
cutoff=0.5 # test dataset, use lower value from range(0,1)
)
barplot(enr.res2d,title='KEGG_2019_Human',)
dotplot(enr.res2d, title='KEGG_2019_Human',)
def run_single(self):
"""run enrichr for one sample"""
# read input file
genes_str=self.parse_input()
# name of analysis or list
description = str(self.descriptions)
gene_set = str(self._gs)
self._logger.info("Connecting to Enrichr Server to get latest library names")
if gene_set in DEFAULT_LIBRARY:
enrichr_library = DEFAULT_LIBRARY
else:
enrichr_library = get_library_name()
if gene_set not in enrichr_library:
sys.stderr.write("%s is not a Enrichr library name\n"%gene_set)
sys.stdout.write("Hint: use get_library_name() to view full list of supported names")
sys.exit(1)
self._logger.info('Analysis name: %s, Enrichr Library: %s'%(description, gene_set))
# enrichr url
ENRICHR_URL = 'http://amp.pharm.mssm.edu/Enrichr/addList'
# payload
payload = {
'list': (None, genes_str),
'description': (None, description)
}
# response
response = requests.post(ENRICHR_URL, files=payload)
if not response.ok:
raise Exception('Error analyzing gene list')
job_id = json.loads(response.text)
self._logger.debug('Job ID:'+ str(job_id))
ENRICHR_URL_A = 'http://amp.pharm.mssm.edu/Enrichr/view?userListId=%s'
user_list_id = job_id['userListId']
response_gene_list = requests.get(ENRICHR_URL_A % str(user_list_id), timeout=None)
# wait for 1s
sleep(1)
if not response_gene_list.ok:
raise Exception('Error getting gene list')
self._logger.info('Submitted gene list:' + str(job_id))
# Get enrichment results
ENRICHR_URL = 'http://amp.pharm.mssm.edu/Enrichr/enrich'
query_string = '?userListId=%s&backgroundType=%s'
# get id data
user_list_id = job_id['userListId']
response = requests.get(ENRICHR_URL + query_string % (str(user_list_id), gene_set))
if not response.ok:
raise Exception('Error fetching enrichment results')
self._logger.debug('Get enrichment results: Job Id:'+ str(job_id))
# Download file of enrichment results
ENRICHR_URL = 'http://amp.pharm.mssm.edu/Enrichr/export'
query_string = '?userListId=%s&filename=%s&backgroundType=%s'
user_list_id = str(job_id['userListId'])
filename = "%s.%s.%s.reports"%(gene_set, description, self.module)
url = ENRICHR_URL + query_string % (user_list_id, filename, gene_set)
# set max retries num =5
s = retry(num=5)
response = s.get(url, stream=True, timeout=None)
self._logger.info('Downloading file of enrichment results: Job Id:'+ str(job_id))
outfile="%s/%s.%s.%s.reports.txt"%(self.outdir, gene_set, description, self.module)
with open(outfile, 'wb') as f:
for chunk in response.iter_content(chunk_size=1024):
if chunk:
f.write(chunk)
self._logger.debug('Results written to: ' + outfile)
# save results
df = read_table(outfile)
self.res2d = df
if self._outdir is None: return
# plotting
if not self.__no_plot:
fig = barplot(df=df, cutoff=self.cutoff,
figsize=self.figsize,
top_term=self.__top_term,
color='salmon',
title='')
if fig is None:
self._logger.warning("Warning: No enrich terms using library %s when cuttoff = %s"%(gene_set, self.cutoff))
else:
fig.savefig(outfile.replace("txt", self.format),
bbox_inches='tight', dpi=300)
self._logger.info('Done.\n')
return
def run(self):
"""run enrichr"""
mkdirs(self.outdir)
#read input file
genes_str = self.parse_input()
# name of analysis or list
description = str(self.descriptions)
#library validaty confirmationi
gene_set = str(self.gene_sets)
#logging start
logger = self._log_init(
module=self.module,
log_level=logging.INFO if self.verbose else logging.WARNING)
logger.info("Connecting to Enrichr Server to get latest library names")
if gene_set in DEFAULT_LIBRARY:
enrichr_library = DEFAULT_LIBRARY
else:
enrichr_library = self.get_libraries()
if gene_set not in enrichr_library:
sys.stderr.write("%s is not a enrichr library name\n" %
gene_set)
sys.stdout.write(
"Hint: use get_library_name() to veiw full list of supported names"
)
sys.exit(1)
logger.info('Analysis name: %s, Enrichr Library: %s' %
(description, gene_set))
## enrichr url
ENRICHR_URL = 'http://amp.pharm.mssm.edu/Enrichr/addList'
# payload
payload = {
'list': (None, genes_str),
'description': (None, description)
}
# response
response = requests.post(ENRICHR_URL, files=payload)
if not response.ok:
raise Exception('Error analyzing gene list')
sleep(1)
job_id = json.loads(response.text)
logger.debug('Job ID:' + str(job_id))
ENRICHR_URL_A = 'http://amp.pharm.mssm.edu/Enrichr/view?userListId=%s'
user_list_id = job_id['userListId']
response_gene_list = requests.get(ENRICHR_URL_A % str(user_list_id),
timeout=None)
if not response_gene_list.ok:
raise Exception('Error getting gene list')
logger.info('Submitted gene list:' + str(job_id))
# Get enrichment results
ENRICHR_URL = 'http://amp.pharm.mssm.edu/Enrichr/enrich'
query_string = '?userListId=%s&backgroundType=%s'
## get id data
user_list_id = job_id['userListId']
response = requests.get(ENRICHR_URL + query_string %
(str(user_list_id), gene_set))
if not response.ok:
raise Exception('Error fetching enrichment results')
logger.debug('Get enrichment results: Job Id:' + str(job_id))
## Download file of enrichment results
ENRICHR_URL = 'http://amp.pharm.mssm.edu/Enrichr/export'
query_string = '?userListId=%s&filename=%s&backgroundType=%s'
user_list_id = str(job_id['userListId'])
filename = "%s.%s.%s.reports" % (gene_set, description, self.module)
url = ENRICHR_URL + query_string % (user_list_id, filename, gene_set)
# set max retries num =5
s = requests.Session()
retries = Retry(total=5,
backoff_factor=0.1,
status_forcelist=[500, 502, 503, 504])
s.mount('http://', HTTPAdapter(max_retries=retries))
response = s.get(url, stream=True, timeout=None)
logger.info('Downloading file of enrichment results: Job Id:' +
str(job_id))
outfile = "%s/%s.%s.%s.reports.txt" % (self.outdir, gene_set,
description, self.module)
with open(outfile, 'wb') as f:
for chunk in response.iter_content(chunk_size=1024):
if chunk:
f.write(chunk)
logger.debug('Results written to: ' + outfile)
#save results
df = read_table(outfile)
self.res2d = df
#plotting
if not self.__no_plot:
fig = barplot(
df=df,
cutoff=self.cutoff,
figsize=self.figsize,
top_term=self.__top_term,
)
if fig is None:
logger.warning(
"Warning: No enrich terms using library %s when cuttoff = %s"
% (gene_set, self.cutoff))
else:
fig.savefig(outfile.replace("txt", self.format),
bbox_inches='tight',
dpi=300)
logger.info('Done.\n')
return
