aboutsummaryrefslogtreecommitdiff
path: root/gn2/wqflask/marker_regression/rqtl_mapping.py
blob: c52add46682058a626dbe8fdd972c180f12a05f3 (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
import csv
import hashlib
import io
import json
import requests
import shutil
from typing import Dict
from typing import List
from typing import Optional
from typing import TextIO
from urllib.parse import urljoin

import numpy as np

from gn3.debug import __pk__

from gn2.base.webqtlConfig import TMPDIR
from gn2.base.trait import create_trait
from gn2.utility.redis_tools import get_redis_conn
from gn2.utility.tools import locate, get_setting, GN3_LOCAL_URL
from gn2.wqflask.database import database_connection


def run_rqtl(trait_name, vals, samples, dataset, pair_scan, mapping_scale, model, method, num_perm, perm_strata_list, do_control, control_marker, manhattan_plot, cofactors):
    """Run R/qtl by making a request to the GN3 endpoint and reading in the output file(s)"""

    pheno_file = write_phenotype_file(trait_name, samples, vals, dataset, cofactors, perm_strata_list)
    if dataset.group.genofile:
        geno_file = locate(dataset.group.genofile, "genotype")
    else:
        geno_file = locate(dataset.group.name + ".geno", "genotype")

    post_data = {
        "pheno_file": pheno_file,
        "geno_file": geno_file,
        "model": model,
        "method": method,
        "nperm": num_perm,
        "scale": mapping_scale
    }

    if pair_scan:
        post_data["pairscan"] = True

    if cofactors:
        covarstruct_file = write_covarstruct_file(cofactors)
        post_data["covarstruct"] = covarstruct_file

    if do_control == "true" and control_marker:
        post_data["control"] = control_marker

    if not manhattan_plot and not pair_scan:
        post_data["interval"] = True
    if cofactors:
        post_data["addcovar"] = True

    if perm_strata_list:
        post_data["pstrata"] = True

    rqtl_output = __pk__(
        "R/qtl or Pair-Scan results",
        requests.post(urljoin(GN3_LOCAL_URL, "api/rqtl/compute"),
                      data=post_data).json())
    if num_perm > 0:
        return rqtl_output['perm_results'], rqtl_output['suggestive'], rqtl_output['significant'], rqtl_output['results']
    else:
        return rqtl_output['results']


def get_hash_of_textio(the_file: TextIO) -> str:
    """Given a StringIO, return the hash of its contents"""

    the_file.seek(0)
    hash_of_file = hashlib.md5(the_file.read().encode()).hexdigest()
    hash_of_file = hash_of_file.replace("/", "_") # Replace / with _ to prevent issue with filenames being translated to directories

    return hash_of_file


def write_covarstruct_file(cofactors: str) -> str:
    """
    Given list of cofactors (as comma-delimited string), write
    a comma-delimited file where the first column consists of cofactor names
    and the second column indicates whether they're numerical or categorical
    """
    trait_datatype_json = None
    with database_connection(get_setting("SQL_URI")) as conn, conn.cursor() as cursor:
        cursor.execute("SELECT value FROM TraitMetadata WHERE type='trait_data_type'")
        trait_datatype_json = json.loads(cursor.fetchone()[0])

    covar_struct_file = io.StringIO()
    writer = csv.writer(covar_struct_file, delimiter="\t", quoting = csv.QUOTE_NONE)
    for cofactor in cofactors.split(","):
        datatype = trait_datatype_json[cofactor] if cofactor in trait_datatype_json else "numerical"
        cofactor_name = cofactor.split(":")[0]
        writer.writerow([cofactor_name, datatype])

    hash_of_file = get_hash_of_textio(covar_struct_file)
    file_path = TMPDIR + hash_of_file + ".csv"

    with open(file_path, "w") as fd:
        covar_struct_file.seek(0)
        shutil.copyfileobj(covar_struct_file, fd)

    return file_path


def write_phenotype_file(trait_name: str,
                         samples: List[str],
                         vals: List,
                         dataset_ob,
                         cofactors: Optional[str] = None,
                         perm_strata_list: Optional[List] = None) -> TextIO:
    """Given trait name, sample list, value list, dataset ob, and optional string
    representing cofactors, return the file's full path/name

    """
    cofactor_data = cofactors_to_dict(cofactors, dataset_ob, samples)

    pheno_file = io.StringIO()
    writer = csv.writer(pheno_file, delimiter="\t", quoting=csv.QUOTE_NONE)

    header_row = ["Samples", trait_name]
    header_row += [cofactor for cofactor in cofactor_data]
    if perm_strata_list:
        header_row.append("Strata")

    writer.writerow(header_row)
    for i, sample in enumerate(samples):
        this_row = [sample]
        if vals[i] != "x":
            this_row.append(str(round(float(vals[i]), 3)))
        else:
            this_row.append("NA")
        for cofactor in cofactor_data:
            this_row.append(cofactor_data[cofactor][i])
        if perm_strata_list:
            this_row.append(perm_strata_list[i])
        writer.writerow(this_row)

    hash_of_file = get_hash_of_textio(pheno_file)
    file_path = TMPDIR + hash_of_file + ".csv"

    with open(file_path, "w") as fd:
        pheno_file.seek(0)
        shutil.copyfileobj(pheno_file, fd)

    return file_path


def cofactors_to_dict(cofactors: str, dataset_ob, samples) -> Dict:
    """Given a string of cofactors, the trait being mapped's dataset ob,
    and list of samples, return cofactor data as a Dict

    """
    cofactor_dict = {}
    if cofactors:
        dataset_ob.group.get_samplelist(redis_conn=get_redis_conn())
        sample_list = dataset_ob.group.samplelist
        for cofactor in cofactors.split(","):
            cofactor_name, cofactor_dataset = cofactor.split(":")
            if cofactor_dataset == dataset_ob.name:
                cofactor_dict[cofactor_name] = []
                trait_ob = create_trait(dataset=dataset_ob,
                                        name=cofactor_name)
                sample_data = trait_ob.data
                for index, sample in enumerate(samples):
                    if sample in sample_data:
                        sample_value = str(round(float(sample_data[sample].value), 3))
                        cofactor_dict[cofactor_name].append(sample_value)
                    else:
                        cofactor_dict[cofactor_name].append("NA")
    return cofactor_dict