aboutsummaryrefslogtreecommitdiff
path: root/scripts/insert_data.py
blob: 53bf4bdd9b005a6bdaa8d0adebef298ae1929948 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
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
"""Insert means/averages or standard-error data into the database."""
import sys
import argparse
from typing import Tuple
from functools import reduce

import MySQLdb as mdb
from redis import Redis
from MySQLdb.cursors import DictCursor

from quality_control.parsing import take
from quality_control.file_utils import open_file
from qc_app.db_utils import database_connection
from qc_app.check_connections import check_db, check_redis

def translate_alias(heading):
    "Translate strain aliases into canonical names"
    translations = {"B6": "C57BL/6J", "D2": "DBA/2J"}
    return translations.get(heading, heading)

def read_file_headings(filepath) -> Tuple[str, ...]:
    "Get the file headings"
    with open_file(filepath) as input_file:
        for line_contents in input_file:
            headings = tuple(
                translate_alias(heading.strip())
                for heading in line_contents.split("\t"))
            break

    return headings

def read_file_contents(filepath):
    "Get the file contents"
    with open_file(filepath) as input_file:
        for line_number, line_contents in enumerate(input_file):
            if line_number == 0:
                continue
            if line_number > 0:
                yield tuple(
                    field.strip() for field in line_contents.split("\t"))

def strains_info(
        dbconn: mdb.Connection, strain_names: Tuple[str, ...],
        speciesid: int) -> dict:
    "Retrieve information for the strains"
    with dbconn.cursor(cursorclass=DictCursor) as cursor:
        query = (
            "SELECT * FROM Strain WHERE Name IN "
            f"({', '.join(['%s']*len(strain_names))}) "
            "AND SpeciesId = %s")
        cursor.execute(query, tuple(strain_names) + (speciesid,))
        return {strain["Name"]: strain for strain in cursor.fetchall()}

def read_datavalues(filepath, headings, strain_info):
    "Read data values from file"
    for row in (
            dict(zip(headings, line))
            for line in read_file_contents(filepath)):
        for sname in headings[1:]:
            yield {
                "ProbeSetId": int(row["ProbeSetID"]),
                "StrainId": strain_info[sname]["Id"],
                "DataValue": float(row[sname])
            }

def last_data_id(dbconn: mdb.Connection) -> int:
    "Get the last id from the database"
    with dbconn.cursor() as cursor:
        cursor.execute("SELECT MAX(Id) FROM ProbeSetData")
        return int(cursor.fetchone()[0])

def check_strains(headings_strains, db_strains):
    "Check strains in headings exist in database"
    from_db = tuple(db_strains.keys())
    not_in_db = tuple(
        strain for strain in headings_strains if strain not in from_db)
    if len(not_in_db) == 0:
        return True

    str_not_in_db = "', '".join(not_in_db)
    print(
        (f"ERROR: The strain(s) '{str_not_in_db}' w(as|ere) not found in the "
         "database."),
        file=sys.stderr)
    sys.exit(1)

def annotationinfo(
        dbconn: mdb.Connection, platformid: int, datasetid: int) -> dict:
    "Get annotation information from the database."
    # This is somewhat slow. Look into optimising the behaviour
    def __organise_annotations__(accm, item):
        names_dict = (
            {**accm[0], item["Name"]: item} if bool(item["Name"]) else accm[0])
        targs_dict = (
            {**accm[1], item["TargetId"]: item}
            if bool(item["TargetId"]) else accm[1])
        return (names_dict, targs_dict)

    query = (
        "SELECT ProbeSet.Name, ProbeSet.ChipId, ProbeSet.TargetId, "
        "ProbeSetXRef.DataId, ProbeSetXRef.ProbeSetFreezeId "
        "FROM ProbeSet INNER JOIN ProbeSetXRef "
        "ON ProbeSet.Id=ProbeSetXRef.ProbeSetId "
        "WHERE ProbeSet.ChipId=%s AND ProbeSetXRef.ProbeSetFreezeId=%s")
    with dbconn.cursor(cursorclass=DictCursor) as cursor:
        cursor.execute(query, (platformid, datasetid))
        annot_dicts = reduce(# type: ignore[var-annotated]
            __organise_annotations__, cursor.fetchall(), ({}, {}))
        return {**annot_dicts[0], **annot_dicts[1]}
    return {}

def insert_means(
        filepath: str, speciesid: int, datasetid: int, dbconn: mdb.Connection,
        rconn: Redis) -> int: # pylint: disable=[unused-argument]
    "Insert the means/averages data into the database"
    print("INSERTING MEANS/AVERAGES DATA.")
    headings = read_file_headings(filepath)
    strains = strains_info(dbconn, headings[1:], speciesid)
    check_strains(headings[1:], strains)
    means_query = (
        "INSERT INTO ProbeSetData "
        "VALUES(%(ProbeSetDataId)s, %(StrainId)s, %(DataValue)s)")
    xref_query = (
        "INSERT INTO ProbeSetXRef(ProbeSetFreezeId, ProbeSetId, DataId) "
        "VALUES (%(ProbeSetFreezeId)s, %(ProbeSetId)s, %(ProbeSetDataId)s)")
    the_means = (
        {"ProbeSetFreezeId": datasetid, "ProbeSetDataId": data_id, **mean}
        for data_id, mean in
        enumerate(
            read_datavalues(filepath, headings, strains),
            start=(last_data_id(dbconn)+1)))
    with dbconn.cursor(cursorclass=DictCursor) as cursor:
        while True:
            means = tuple(take(the_means, 1000))
            if not bool(means):
                break
            print(
                f"\nEXECUTING QUERIES:\n\t* {means_query}\n\t* {xref_query}\n"
                f"with parameters\n\t{means}")
            cursor.executemany(means_query, means)
            cursor.executemany(xref_query, means)
    return 0

def insert_se(# pylint: disable = [too-many-arguments]
        filepath: str, speciesid: int, platformid: int, datasetid: int,
        dbconn: mdb.Connection, rconn: Redis) -> int: # pylint: disable=[unused-argument]
    "Insert the standard-error data into the database"
    print("INSERTING STANDARD ERROR DATA...")
    headings = read_file_headings(filepath)
    strains = strains_info(dbconn, headings[1:], speciesid)
    check_strains(headings[1:], strains)
    se_query = (
        "INSERT INTO ProbeSetSE "
        "VALUES(%(DataId)s, %(StrainId)s, %(DataValue)s)")
    annotations = annotationinfo(dbconn, platformid, datasetid)
    if not bool(annotations):
        print(
            (f"ERROR: No annotations found for platform {platformid} and "
             f"dataset {datasetid}. Quiting!"),
            file=sys.stderr)
        return 1

    se_values = (
        {"DataId": annotations[str(item["ProbeSetId"])]["DataId"], **item}
        for item in read_datavalues(filepath, headings, strains))
    with dbconn.cursor(cursorclass=DictCursor) as cursor:
        while True:
            serrors = tuple(take(se_values, 1000))
            if not bool(serrors):
                break
            print(
                f"\nEXECUTING QUERY:\n\t* {se_query}\n\tWITH PARAMETERS:\n\t"
                f"{serrors}")
            cursor.executemany(se_query, serrors)
    return 0

if __name__ == "__main__":
    def cli_args():
        "Compute the CLI arguments"
        parser = argparse.ArgumentParser(
            prog="InsertData", description=(
                "Script to insert data from an 'averages' file into the "
                "database."))
        parser.add_argument(
            "filetype", help="type of data to insert.",
            choices=("average", "standard-error"))
        parser.add_argument(
            "filepath", help="path to the file with the 'averages' data.")
        parser.add_argument(
            "speciesid", help="Identifier for the species in the database.",
            type=int)
        parser.add_argument(
            "platformid", help="Identifier for the platform in the database.",
            type=int)
        parser.add_argument(
            "datasetid", help="Identifier for the dataset in the database.",
            type=int)
        parser.add_argument(
            "database_uri",
            help="URL to be used to initialise the connection to the database")
        parser.add_argument(
            "redisuri",
            help="URL to initialise connection to redis",
            default="redis:///")

        args = parser.parse_args()
        check_db(args.database_uri)
        check_redis(args.redisuri)
        return args

    insert_fns = {
        "average": insert_means,
        "standard-error": insert_se
    }

    extract_args = {
        "average": lambda args, dbconn, rconn: (
            args.filepath, args.speciesid, args.datasetid, dbconn, rconn),
        "standard-error": lambda args, dbconn, rconn: (
            args.filepath, args.speciesid, args.platformid, args.datasetid,
            dbconn, rconn),
    }

    def main():
        "Main entry point"
        args = cli_args()
        with Redis.from_url(args.redisuri, decode_responses=True) as rconn:
            with database_connection(args.database_uri) as dbconn:
                return insert_fns[args.filetype](
                    *extract_args[args.filetype](args, dbconn, rconn))

        return 2

    sys.exit(main())