Skip to content

prepare_genome

ensembl.brc4.runnable.prepare_genome

prepare_genome

Bases: BaseRunnable

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
class prepare_genome(eHive.BaseRunnable):
    def param_defaults(self):
        return {"ensembl_mode": False, "db_prefix": ""}

    def run(self):
        manifest_path = self.param_required("manifest")

        errors = []

        manifest = self.get_manifest(manifest_path)
        genome = self.get_genome(manifest)
        self.update_accession(genome)
        self.update_assembly_name(genome)

        print(manifest)
        print("")
        print(genome)

        db_name = self.make_db_name(genome)
        print("db_name = %s" % db_name)
        self.param("db_name", db_name)

        # FLOW
        self.dataflow(
            {
                "db_name": db_name,
                "manifest_data": manifest,
                "genome_data": genome,
                "species": genome["species"]["production_name"],
                "has_gff3": (1 if "gff3" in manifest and manifest["gff3"] else 0),
            },
            2,
        )

        # DB metadata for registry
        self.dataflow(
            {
                "db_data": {
                    "db_name": db_name,
                    "genome_data": genome,
                }
            },
            3,
        )

    def get_manifest(self, manifest_path):
        with open(manifest_path) as manifest_file:
            manifest = json.load(manifest_file)

            # Use dir name from the manifest
            for name in manifest:
                if "file" in manifest[name]:
                    file_name = manifest[name]["file"]
                    manifest[name] = path.join(path.dirname(manifest_path), file_name)
                else:
                    for f in manifest[name]:
                        if "file" in manifest[name][f]:
                            file_name = manifest[name][f]["file"]
                            manifest[name][f] = path.join(path.dirname(manifest_path), file_name)
            return manifest

    def get_genome(self, manifest):
        # Get genome metadata
        if "genome" in manifest:
            genome_path = manifest["genome"]
            print(genome_path)

            with open(genome_path) as genome_file:
                return json.load(genome_file)

    def make_db_name(self, genome):
        prod_name = self.make_production_name(genome)
        prod_name = prod_name.replace(".", "")
        ensembl_version = str(self.param_required("ensembl_version"))
        release = str(self.param_required("release"))
        assembly_version = str(self.get_assembly_version(genome))
        db_prefix = self.param("db_prefix")

        # Check the db name is shorter than the limit: reduce it to "Genus sp"
        db_prod_name = prod_name
        max_prod_length = (
            64 - len("core") - len(str(release)) - len(str(ensembl_version)) - len(str(assembly_version)) - 4
        )
        if db_prefix:
            max_prod_length = max_prod_length - len(db_prefix) - 1
        if len(db_prod_name) > max_prod_length:
            print("DB name is too long! Trying to shorten it...")
            (genus_str, species_str, gca_str) = db_prod_name.split("_")
            db_prod_name = "_".join([genus_str, "sp", gca_str])

            # Still too long? Die
            if len(db_prod_name) > max_prod_length:
                raise Exception(
                    "Can't use the db name for %s (len = %d, max = %d)"
                    % (db_prod_name, len(db_prod_name), max_prod_length)
                )
            else:
                prod_name = db_prod_name

        # Store this production name
        genome["species"]["production_name"] = prod_name

        name_parts = [prod_name, "core", release, ensembl_version, assembly_version]

        db_name = "_".join(name_parts)
        db_name = db_name.replace(".", "")
        self.check_db_name_format(db_name)

        if db_prefix:
            db_name = db_prefix + "_" + db_name

        return db_name

    def make_production_name(self, genome):
        if "species" in genome:
            genome_sp = genome["species"]
            if "production_name" in genome_sp:
                prod_name = genome_sp["production_name"]
            else:
                # Create prod name from taxonomy
                if "taxonomy_id" in genome_sp:
                    taxon_id = genome_sp["taxonomy_id"]
                    scientific_name = self.get_scientific_name(taxon_id)

                    # If what we get is not a species name (only 1 word), use the input name
                    if " " not in scientific_name:
                        print("Use input scientific name")
                        scientific_name = genome_sp["scientific_name"]

                    # Name in bracket means that it may be renamed by the community
                    # But we don't want those...
                    if "[" in scientific_name:
                        scientific_name = scientific_name.replace("[", "").replace("]", "")

                    if "-" in scientific_name:
                        scientific_name = scientific_name.replace("-", "")

                    # Only keep the first 2 words
                    split_name = scientific_name.split(" ")
                    genus = split_name[0].lower()
                    species = split_name[1].lower()

                    # Special case
                    if species == "sp.":
                        species = "sp"

                    # Check for special characters
                    if not re.match(r"^[a-zA-Z0-9]+$", genus):
                        raise Exception("The genus has special characters: " + genus)
                    if re.match(r"^[a-zA-Z0-9]$", species):
                        raise Exception("The species has special characters: " + species)
                    print(genus + " " + species)

                    # Production name, with INSDC accession for uniqueness
                    accession = genome["assembly"]["accession"]
                    if not accession or accession == "":
                        raise Exception("The INSDC accession is needed")
                    accession = re.sub("\.\d+$", "", accession).replace("_", "")
                    accession = accession.lower()
                    prod_name = genus + "_" + species + "_" + accession
                else:
                    raise Exception("Can't find taxonomy id for genome: %s" % genome)
            return prod_name
        else:
            raise Exception("Can't make production name for genome: %s" % genome)

    def get_scientific_name(self, taxon_id):
        host = self.param("taxonomy_host")
        user = self.param("taxonomy_user")
        port = self.param("taxonomy_port")
        dbname = self.param("taxonomy_dbname")

        # Get connector to taxon db
        mydb = mysql.connector.connect(host=host, user=user, port=port, database=dbname)
        cur = mydb.cursor()

        # Get scientific name
        select_st = "SELECT name FROM ncbi_taxa_name WHERE name_class=%s and taxon_id=%s"
        data = ("scientific name", taxon_id)
        cur.execute(select_st, data)
        row = cur.fetchone()

        if row:
            for s in row:
                return s
        else:
            return

    def get_assembly_version(self, genome):
        if "assembly" in genome:
            assembly = genome["assembly"]
            if "version" in assembly:
                aversion = assembly["version"]

                if str(aversion) == "0":
                    raise Exception("Assembly version cannot be 0")

                if isinstance(aversion, int):
                    return aversion
                else:
                    try:
                        aversion = int(aversion)
                        genome["assembly"]["version"] = aversion
                        return aversion
                    except:
                        raise Exception("Assembly version is not an integer: %s" % aversion)
            elif "name" in assembly:
                # Get last number at the end of the assembly name
                matches = re.match("\w+?(\d+?)$", assembly["name"])
                if matches:
                    aversion = matches.group(1)
                    genome["assembly"]["version"] = aversion
                    return aversion
                else:
                    raise Exception("No assembly version found in %s" % str(genome))
            else:
                raise Exception("Can't get assembly version from name in from %s" % str(genome))
        else:
            raise Exception("Can't get assembly data in genome %s" % str(genome))

    def check_db_name_format(self, db_name):
        match = re.match("[a-z]+_[a-z]+(_[A-z0-9]+){0,2}_core_\d+_\d+_\d+$", db_name)

        if not match:
            raise Exception(
                "Generated DB name is not the right format: %s (in %s)"
                % (db_name, self.param_required("manifest"))
            )

    def update_assembly_name(self, data):
        if data and "assembly" in data:
            asm = data["assembly"]
            if "name" not in asm:
                if "accession" not in asm:
                    raise Exception("no accession or name in genome_data/assembly")
                acc = asm["accession"].replace("_", "").replace(".", "v")
                asm["name"] = acc
                print('using "%s" as assembly.name' % (data["assembly"]["name"]), file=sys.stderr)
        else:
            raise Exception("no 'assembly' data provided in genome_data")

    def update_accession(self, data):
        if data and "assembly" in data:
            asm = data["assembly"]
            if "accession" in asm and asm["accession"].startswith("GCF_"):
                ### CHANGE TO GCA
                data["assembly"]["accession"] = data["assembly"]["accession"].replace("GCF_", "GCA_")
                print('using "%s" as assembly.accession' % (data["assembly"]["accession"]), file=sys.stderr)
        else:
            raise Exception("no 'assembly' data provided in genome_data")

check_db_name_format(db_name)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
247
248
249
250
251
252
253
254
def check_db_name_format(self, db_name):
    match = re.match("[a-z]+_[a-z]+(_[A-z0-9]+){0,2}_core_\d+_\d+_\d+$", db_name)

    if not match:
        raise Exception(
            "Generated DB name is not the right format: %s (in %s)"
            % (db_name, self.param_required("manifest"))
        )

get_assembly_version(genome)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
def get_assembly_version(self, genome):
    if "assembly" in genome:
        assembly = genome["assembly"]
        if "version" in assembly:
            aversion = assembly["version"]

            if str(aversion) == "0":
                raise Exception("Assembly version cannot be 0")

            if isinstance(aversion, int):
                return aversion
            else:
                try:
                    aversion = int(aversion)
                    genome["assembly"]["version"] = aversion
                    return aversion
                except:
                    raise Exception("Assembly version is not an integer: %s" % aversion)
        elif "name" in assembly:
            # Get last number at the end of the assembly name
            matches = re.match("\w+?(\d+?)$", assembly["name"])
            if matches:
                aversion = matches.group(1)
                genome["assembly"]["version"] = aversion
                return aversion
            else:
                raise Exception("No assembly version found in %s" % str(genome))
        else:
            raise Exception("Can't get assembly version from name in from %s" % str(genome))
    else:
        raise Exception("Can't get assembly data in genome %s" % str(genome))

get_genome(manifest)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
88
89
90
91
92
93
94
95
def get_genome(self, manifest):
    # Get genome metadata
    if "genome" in manifest:
        genome_path = manifest["genome"]
        print(genome_path)

        with open(genome_path) as genome_file:
            return json.load(genome_file)

get_manifest(manifest_path)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
def get_manifest(self, manifest_path):
    with open(manifest_path) as manifest_file:
        manifest = json.load(manifest_file)

        # Use dir name from the manifest
        for name in manifest:
            if "file" in manifest[name]:
                file_name = manifest[name]["file"]
                manifest[name] = path.join(path.dirname(manifest_path), file_name)
            else:
                for f in manifest[name]:
                    if "file" in manifest[name][f]:
                        file_name = manifest[name][f]["file"]
                        manifest[name][f] = path.join(path.dirname(manifest_path), file_name)
        return manifest

get_scientific_name(taxon_id)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
def get_scientific_name(self, taxon_id):
    host = self.param("taxonomy_host")
    user = self.param("taxonomy_user")
    port = self.param("taxonomy_port")
    dbname = self.param("taxonomy_dbname")

    # Get connector to taxon db
    mydb = mysql.connector.connect(host=host, user=user, port=port, database=dbname)
    cur = mydb.cursor()

    # Get scientific name
    select_st = "SELECT name FROM ncbi_taxa_name WHERE name_class=%s and taxon_id=%s"
    data = ("scientific name", taxon_id)
    cur.execute(select_st, data)
    row = cur.fetchone()

    if row:
        for s in row:
            return s
    else:
        return

make_db_name(genome)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
def make_db_name(self, genome):
    prod_name = self.make_production_name(genome)
    prod_name = prod_name.replace(".", "")
    ensembl_version = str(self.param_required("ensembl_version"))
    release = str(self.param_required("release"))
    assembly_version = str(self.get_assembly_version(genome))
    db_prefix = self.param("db_prefix")

    # Check the db name is shorter than the limit: reduce it to "Genus sp"
    db_prod_name = prod_name
    max_prod_length = (
        64 - len("core") - len(str(release)) - len(str(ensembl_version)) - len(str(assembly_version)) - 4
    )
    if db_prefix:
        max_prod_length = max_prod_length - len(db_prefix) - 1
    if len(db_prod_name) > max_prod_length:
        print("DB name is too long! Trying to shorten it...")
        (genus_str, species_str, gca_str) = db_prod_name.split("_")
        db_prod_name = "_".join([genus_str, "sp", gca_str])

        # Still too long? Die
        if len(db_prod_name) > max_prod_length:
            raise Exception(
                "Can't use the db name for %s (len = %d, max = %d)"
                % (db_prod_name, len(db_prod_name), max_prod_length)
            )
        else:
            prod_name = db_prod_name

    # Store this production name
    genome["species"]["production_name"] = prod_name

    name_parts = [prod_name, "core", release, ensembl_version, assembly_version]

    db_name = "_".join(name_parts)
    db_name = db_name.replace(".", "")
    self.check_db_name_format(db_name)

    if db_prefix:
        db_name = db_prefix + "_" + db_name

    return db_name

make_production_name(genome)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
def make_production_name(self, genome):
    if "species" in genome:
        genome_sp = genome["species"]
        if "production_name" in genome_sp:
            prod_name = genome_sp["production_name"]
        else:
            # Create prod name from taxonomy
            if "taxonomy_id" in genome_sp:
                taxon_id = genome_sp["taxonomy_id"]
                scientific_name = self.get_scientific_name(taxon_id)

                # If what we get is not a species name (only 1 word), use the input name
                if " " not in scientific_name:
                    print("Use input scientific name")
                    scientific_name = genome_sp["scientific_name"]

                # Name in bracket means that it may be renamed by the community
                # But we don't want those...
                if "[" in scientific_name:
                    scientific_name = scientific_name.replace("[", "").replace("]", "")

                if "-" in scientific_name:
                    scientific_name = scientific_name.replace("-", "")

                # Only keep the first 2 words
                split_name = scientific_name.split(" ")
                genus = split_name[0].lower()
                species = split_name[1].lower()

                # Special case
                if species == "sp.":
                    species = "sp"

                # Check for special characters
                if not re.match(r"^[a-zA-Z0-9]+$", genus):
                    raise Exception("The genus has special characters: " + genus)
                if re.match(r"^[a-zA-Z0-9]$", species):
                    raise Exception("The species has special characters: " + species)
                print(genus + " " + species)

                # Production name, with INSDC accession for uniqueness
                accession = genome["assembly"]["accession"]
                if not accession or accession == "":
                    raise Exception("The INSDC accession is needed")
                accession = re.sub("\.\d+$", "", accession).replace("_", "")
                accession = accession.lower()
                prod_name = genus + "_" + species + "_" + accession
            else:
                raise Exception("Can't find taxonomy id for genome: %s" % genome)
        return prod_name
    else:
        raise Exception("Can't make production name for genome: %s" % genome)

param_defaults()

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
28
29
def param_defaults(self):
    return {"ensembl_mode": False, "db_prefix": ""}

run()

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
def run(self):
    manifest_path = self.param_required("manifest")

    errors = []

    manifest = self.get_manifest(manifest_path)
    genome = self.get_genome(manifest)
    self.update_accession(genome)
    self.update_assembly_name(genome)

    print(manifest)
    print("")
    print(genome)

    db_name = self.make_db_name(genome)
    print("db_name = %s" % db_name)
    self.param("db_name", db_name)

    # FLOW
    self.dataflow(
        {
            "db_name": db_name,
            "manifest_data": manifest,
            "genome_data": genome,
            "species": genome["species"]["production_name"],
            "has_gff3": (1 if "gff3" in manifest and manifest["gff3"] else 0),
        },
        2,
    )

    # DB metadata for registry
    self.dataflow(
        {
            "db_data": {
                "db_name": db_name,
                "genome_data": genome,
            }
        },
        3,
    )

update_accession(data)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
268
269
270
271
272
273
274
275
276
def update_accession(self, data):
    if data and "assembly" in data:
        asm = data["assembly"]
        if "accession" in asm and asm["accession"].startswith("GCF_"):
            ### CHANGE TO GCA
            data["assembly"]["accession"] = data["assembly"]["accession"].replace("GCF_", "GCA_")
            print('using "%s" as assembly.accession' % (data["assembly"]["accession"]), file=sys.stderr)
    else:
        raise Exception("no 'assembly' data provided in genome_data")

update_assembly_name(data)

Source code in src/python/ensembl/brc4/runnable/prepare_genome.py
256
257
258
259
260
261
262
263
264
265
266
def update_assembly_name(self, data):
    if data and "assembly" in data:
        asm = data["assembly"]
        if "name" not in asm:
            if "accession" not in asm:
                raise Exception("no accession or name in genome_data/assembly")
            acc = asm["accession"].replace("_", "").replace(".", "v")
            asm["name"] = acc
            print('using "%s" as assembly.name' % (data["assembly"]["name"]), file=sys.stderr)
    else:
        raise Exception("no 'assembly' data provided in genome_data")