Skip to content

Commit

Permalink
added generation of bigwig URLs
Browse files Browse the repository at this point in the history
  • Loading branch information
@evanizer8
evanizer8 committed Oct 28, 2024
1 parent a083d8f commit 0fd4094
Showing 1 changed file with 63 additions and 4 deletions.
67 changes: 63 additions & 4 deletions scripts/process_collections.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -156,7 +156,7 @@ def process_collections(self, cmds_only, mode):
if mode == "jbrowse": # for jbrowse
cmd = f"jbrowse add-assembly -n {name} --out {os.path.abspath(self.out_dir)}/ -t bgzipFasta --force"
cmd += f' --displayName "{genus.capitalize()} {species} {infraspecies} V{version.replace("gnm", "")} {collection_type.capitalize()}" {url}'
print("url: ", url)
# print("url: ", url)
elif mode == "blast": # for blast
cmd = f"set -o pipefail -o errexit -o nounset; curl {url} | gzip -dc" # retrieve genome and decompress
cmd += f'| makeblastdb -parse_seqids -out {self.out_dir}/{name} -hash_index -dbtype nucl -title "{genus.capitalize()} {species} {infraspecies} V{version.replace("gnm", "")} {collection_type.capitalize()}"'
Expand Down Expand Up @@ -408,6 +408,7 @@ def add_collections(self, collection_type, genus, species):
else: # fai file could not be accessed
logger.error(f"No fai file for: {url}")
sys.exit(1)

linear_session = { # LinearGenomeView object for JBrowse2
"views": [
{
Expand All @@ -424,13 +425,17 @@ def add_collections(self, collection_type, genus, species):
]
}
linear_url = f"{self.jbrowse_url}/?config=config.json&session=spec-{linear_session}" # build the URL for the resource

linear_data = {
"name": f"JBrowse2 {lookup}",
"URL": str(linear_url).replace(
"'", "%22"
), # url encode for .yml file and Jekyll linking
"description": "JBrowse2 Linear Genome View",
} # the object that will be written into the .yml file

logger.debug(f"linear data for assembly: {linear_data} \n")

if strain_lookup not in self.infraspecies_resources:
self.infraspecies_resources[strain_lookup] = (
[]
Expand Down Expand Up @@ -610,7 +615,7 @@ def add_collections(self, collection_type, genus, species):
) # add data for later writing in resources
###
elif collection_type == "expression": # add parent expr files

ref=""
# Synteny after the new changes. Parent is a tuple with both genome_main files
checksum_url = (
f"{self.datastore_url}{collection_dir}CHECKSUM.{parts[1]}.md5"
Expand All @@ -632,11 +637,13 @@ def add_collections(self, collection_type, genus, species):
fields = line.split()
if fields: # process if fields exists
if fields[1].endswith("bw"): # get bw file
ref = ""
stop = 0
bw_lookup = fields[1].replace(
"./", ""
) # get bw file to load will start with ./
logger.debug(bw_lookup)
bw_url = f"{self.datastore_url}{collection_dir}{bw_lookup}" # where the paf file is in the datastore
bw_url = f"{self.datastore_url}{collection_dir}{bw_lookup}" # where the bigwig file is in the datastore
bw_parts = bw_lookup.split(
"."
) # split the bw file name into parts delimited by '.'
Expand All @@ -655,6 +662,58 @@ def add_collections(self, collection_type, genus, species):
"taxid": 0,
}
logger.debug(self.files[collection_type][bw_lookup])

#url = f"{self.datastore_url}{collection_dir}{parts[0]}.{parts[1]}.genome_main.fna.gz" # genome_main in datastore_url
url = self.files['genomes'][parent]['url']
fai_url = f"{url}.fai" # get fai file for jbrowse session construction
fai_response = self.get_remote(
fai_url
) # get fai file to build loc from
if fai_response: # fai SUCCESS 200
(ref, stop) = fai_response.split("\n")[0].split()[
:2
] # fai field 1\s+2. field 1 is sequence_id field 2 is length
logger.debug(f"{ref},{stop}")
else: # fai file could not be accessed
logger.error(f"No fai file for: {url}")
sys.exit(1)


linear_session = { # LinearGenomeView object for JBrowse2
"views": [
{
"assembly": parent,
#sequence is currently hardcoded, don't know how "ref" works for genomes
"loc": f"{ref}:1-{stop}", # JBrowse2 does not allow null loc
"type": "LinearGenomeView",
"tracks": [".".join(bw_lookup.split(".")[:-1])]
#["glyma.Wm82.gnm6.ann1.expr.mixed.Kour_Boone_2014.Clark_defective"]
# " gff3tabix_genes " ,
# " volvox_filtered_vcf " ,
# " volvox_microarray " ,
# " volvox_cram "
# ]
}
]
}
linear_url = f"{self.jbrowse_url}/?config=config.json&session=spec-{linear_session}" # build the URL for the resource

linear_data = {
"name": f"JBrowse2 {parent}",
"URL": str(linear_url).replace(
"'", "%22"
), # url encode for .yml file and Jekyll linking
"description": "JBrowse2 Linear Genome View",
} # the object that will be written into the .yml file

if strain_lookup not in self.infraspecies_resources:
self.infraspecies_resources[strain_lookup] = (
[]
) # initialize infraspecies list within species
if self.jbrowse_url: # dont add data if no jbrowse url set
self.infraspecies_resources[strain_lookup].append(linear_data)

logger.debug(f"linear data for bw: {linear_data} \n")
###

readme_url = f"{self.datastore_url}/{collection_dir}README.{name}.yml" # species collection readme
Expand Down Expand Up @@ -828,7 +887,7 @@ def parse_collections(
"""Retrieve and output collections for jekyll site"""
if from_github: # set to None if empty dir
self.from_github = os.path.abspath(from_github)
print(f"THIS IS GITHUB: {self.from_github}")
self.logger.debug(f"THIS IS GITHUB: {self.from_github}")
taxon_list = yaml.load(
open(target, "r", encoding="utf-8").read(), Loader=yaml.FullLoader
) # load taxon list
Expand Down

0 comments on commit 0fd4094

Please sign in to comment.