Merge pull request #130 from rigdenlab/development

Development

Author: FilomenoSanchez

app.py

@@ -203,6 +203,7 @@ def recover_account(n_clicks, username, email, secret, password_1, password_2):
 
     return app_utils.recover_account(username, email, secret, password_1, password_2, app.logger)
 
+
 @app.callback([Output('invalid-create-user-collapse', 'is_open'),
                Output('create-user-modal-div', 'children'),
                Output('create-username-input', 'value'),

components/__init__.py

@@ -94,6 +94,12 @@ def TutorialFourModal(*args, **kwargs):
     return TutorialFourModal(*args, **kwargs)
 
 
+def TutorialFiveModal(*args, **kwargs):
+    from components.modals import TutorialFiveModal
+
+    return TutorialFiveModal(*args, **kwargs)
+
+
 def GdprAgreementCheckbox(*args, **kwargs):
     from components.inputgroups import GdprAgreementCheckbox
 
@@ -454,6 +460,12 @@ def NoAdditionalTracksCard(*args, **kwargs):
     return NoAdditionalTracksCard(*args, **kwargs)
 
 
+def MismatchDatasetModal(*args, **kwargs):
+    from components.modals import MismatchDatasetModal
+
+    return MismatchDatasetModal(*args, **kwargs)
+
+
 def ContactDisplayControlCard(*args, **kwargs):
     from components.cards import DisplayControlCard
 

components/listgrpoups.py

@@ -93,7 +93,8 @@ def TutorialList():
             TutorialItem(idx=1, name='Creating your first plot'),
             TutorialItem(idx=2, name='Compare a contact prediction with a PDB file'),
             TutorialItem(idx=3, name='Storing, loading and sharing a session'),
-            TutorialItem(idx=4, name='Residue-Residue distance predictions')
+            TutorialItem(idx=4, name='Residue-Residue distance predictions'),
+            #TutorialItem(idx=5, name='Video tutorial')
         ], style={'width': '75%'}
         ), justify='center', align='center')
 

components/modals.py

@@ -16,12 +16,23 @@ def MismatchModal(*args):
     ], id='mismatch-modal', is_open=True)
 
 
+def MismatchDatasetModal(fname, seq_fname):
+    return dbc.Modal([
+        ModalHeader("Mismatch Detected"),
+        dbc.ModalBody([
+            html.P("""We were unable to match the sequence at {} with the data at the file {}. Please 
+                ensure that the file you attempt to upload corresponds with the protein sequence 
+                in the provided FASTA file.""".format(seq_fname, fname), style={'text-align': "justify"}),
+        ])
+    ], id='mismatch-dataset-modal', is_open=True)
+
+
 def MismatchSequenceModal(*args):
     return dbc.Modal([
         ModalHeader("Sequence Mismatch"),
         dbc.ModalBody([
-            html.P("""We were unable to match the uploaded sequence with contact maps in the following files. 
-            Please ensure that the provided the sequence corresponds with the structure in these contact maps.""",
+            html.P("""We were unable to match the uploaded sequence with datasets in the following files. 
+            Please ensure that the provided sequence corresponds with the structure described in these datasets.""",
                    style={'text-align': "justify"}),
             html.Ul([html.Li('File: %s' % arg) for arg in args], id='mismatched-maps-div')
         ])
@@ -541,6 +552,23 @@ def TutorialFourModal():
         autoFocus=True)
 
 
+def TutorialFiveModal():
+    return dbc.Modal([
+        dbc.ModalHeader('Tutorial 5: Video Tutorial'),
+        dbc.ModalBody([
+            'Below there is a short video tutorial with an overview of the main ConPlot features. ',
+            html.Br(),
+            html.Br(),
+            html.Iframe(width=1100, height=630, src=UrlIndex.YOUTUBE_EMBED.value,
+                        style={'frameborder': 0, 'allow': "accelerometer; autoplay; clipboard-write; "
+                                                          "encrypted-media; gyroscope; picture-in-picture"}),
+            html.Br(),
+            'If you cannot see the video click ', html.A(html.U('here'), href=UrlIndex.YOUTUBE_LINK.value), '.'
+        ])
+    ], id={'type': 'tutorial-modal', 'index': 5}, is_open=False, size='xl', scrollable=True, centered=True,
+        autoFocus=True)
+
+
 def RedisConnectionErrorModal():
     return dbc.Modal([
         ModalHeader("Redis connection error"),
@@ -679,6 +707,7 @@ def FailureRecoverAccount():
         ),
     ], id='fail-recovery-modal', is_open=True)
 
+
 def InvalidPasswordRecoverAccount():
     return dbc.Modal([
         dbc.ModalHeader(

layouts/help.py

@@ -21,6 +21,7 @@ def Body(cache):
             components.TutorialTwoModal(),
             components.TutorialThreeModal(),
             components.TutorialFourModal(),
+            #components.TutorialFiveModal(),
             components.CustomFormatDescriptionModal(),
             dbc.Row([
                 dbc.Col([

layouts/home.py

@@ -90,7 +90,7 @@ def Body():
                             html.P('If you found ConPlot useful for your work and you would like to cite us '
                                    'please use the following reference:',
                                    style={"font-size": "120%", "text-align": "justify"}),
-                            html.I('Sánchez Rodríguez, et al., ConPlot: Web-based application for the visualisation '
+                            html.I('Sánchez Rodríguez F. et al., ConPlot: Web-based application for the visualisation '
                                    'of protein contact maps integrated with other data, Bioinformatics, 2021.',
                                    style={"font-size": "120%", "text-align": "justify"}),
                             html.Br(),

loaders/loader.py

@@ -10,15 +10,14 @@ def Loader(raw_file, input_format):
 
     if raw_file is not None:
         try:
-            decoded = decode_raw_file(raw_file)
-            if input_format in ContactInformationFormats.__members__ and input_format != ContactInformationFormats.CCMPRED.name \
-                    and input_format != ContactInformationFormats.PDB.name:
+            if input_format != ContactInformationFormats.trROSETTA_NPZ.name:
+                decoded = decode_raw_file(raw_file)
                 data_raw = ParserFormats.__dict__[input_format](decoded, input_format)
             else:
-                data_raw = ParserFormats.__dict__[input_format](decoded)
+                data_raw = ParserFormats.__dict__[input_format](raw_file, input_format)
             data = compress_data(data_raw)
         except (InvalidFormat, UnicodeDecodeError) as e:
             data = None
             invalid = True
 
-    return data, invalid
+    return data, invalid

parsers/__init__.py

@@ -25,10 +25,10 @@ def guess_psipred_format(*args, **kwargs):
     return guess_psipred_format(*args, **kwargs)
 
 
-def DistogramParser(*args, **kwargs):
-    from parsers.distogramparser import DistogramParser
+def CASPRR2Parser(*args, **kwargs):
+    from parsers.casprr2parser import CASPRR2Parser
 
-    return DistogramParser(*args, **kwargs)
+    return CASPRR2Parser(*args, **kwargs)
 
 
 def TopconsParser(*args, **kwargs):
@@ -73,6 +73,12 @@ def CustomParser(*args, **kwargs):
     return CustomParser(*args, **kwargs)
 
 
+def NpzParser(*args, **kwargs):
+    from parsers.npzparser import NpzParser
+
+    return NpzParser(*args, **kwargs)
+
+
 class ParserFormats(Enum):
     TOPCONS = TopconsParser
     CONSURF = ConsurfParser
@@ -97,7 +103,8 @@ class ParserFormats(Enum):
     CCMPRED = CCMpredParser
     COLSTATS = CCMpredParser
     PDB = PDBParser
-    CASPRR_MODE_2 = DistogramParser
+    CASPRR_MODE_2 = CASPRR2Parser
+    trROSETTA_NPZ = NpzParser
 
 
 class ContactInformationFormats(Enum):
@@ -122,6 +129,7 @@ class ContactInformationFormats(Enum):
     MAPALIGN = 18
     ALEIGEN = 19
     PDB = 20
+    trROSETTA_NPZ = 21
 
 
 class ContactMapFormats(Enum):
@@ -152,6 +160,7 @@ class StructuralInformationFormats(Enum):
 
 class DistanceInformationFormats(Enum):
     CASPRR_MODE_2 = 1
+    trROSETTA_NPZ = 2
 
 
 class MembraneStates(Enum):

parsers/distogramparser.py renamed to parsers/casprr2parser.py

@@ -1,9 +1,8 @@
-from operator import itemgetter
-from utils import unique_by_key
+from utils import get_unique_distances
 from utils.exceptions import InvalidFormat
 
 
-def DistogramParser(input, input_format=None):
+def CASPRR2Parser(input, input_format=None):
     contents = input.split('\n')
     output = []
     res_1_idx = 0
@@ -32,10 +31,7 @@ def DistogramParser(input, input_format=None):
             output.append((tuple(contact[:2]), *contact[2:]))
 
     if not output:
-        raise InvalidFormat('Unable to parse contacts')
+        raise InvalidFormat('Unable to parse CASPRR_MODE_2 file')
     else:
-        unique_contacts = unique_by_key(output, key=itemgetter(0))
-        output = [(*contact[0], *contact[1:]) for contact in unique_contacts]
-        output = sorted(output, key=itemgetter(2), reverse=True)
-        output.append('DISTO')
-        return output
+        unique_contacts = get_unique_distances(output)
+        return unique_contacts

parsers/ccmpredparser.py

@@ -1,9 +1,8 @@
-from operator import itemgetter
 from utils.exceptions import InvalidFormat
-from utils import unique_by_key
+from utils import get_unique_contacts
 
 
-def CCMpredParser(input):
+def CCMpredParser(input, input_format=None):
     contents = input.split('\n')
 
     output = []
@@ -31,7 +30,5 @@ def CCMpredParser(input):
     if not output:
         raise InvalidFormat('Unable to parse contacts')
     else:
-        unique_contacts = unique_by_key(output, key=itemgetter(0))
-        output = [(*contact[0], contact[1]) for contact in unique_contacts]
-        output = sorted(output, key=itemgetter(2), reverse=True)
-        return output
+        unique_contacts = get_unique_contacts(output)
+        return unique_contacts

parsers/consurfparser.py

@@ -1,7 +1,7 @@
 from utils.exceptions import InvalidFormat
 
 
-def ConsurfParser(input):
+def ConsurfParser(input, input_format=None):
     contents = input.split('\n')
     output = []
 

parsers/contactparser.py

@@ -1,7 +1,6 @@
 from enum import Enum
-from operator import itemgetter
 import re
-from utils import unique_by_key
+from utils import get_unique_contacts
 from utils.exceptions import InvalidFormat
 
 
@@ -138,7 +137,5 @@ def ContactParser(input, input_format):
     if not output:
         raise InvalidFormat('Unable to parse contacts')
     else:
-        unique_contacts = unique_by_key(output, key=itemgetter(0))
-        output = [(*contact[0], contact[1]) for contact in unique_contacts]
-        output = sorted(output, key=itemgetter(2), reverse=True)
-        return output
+        unique_contacts = get_unique_contacts(output)
+        return unique_contacts

parsers/customparser.py

@@ -2,7 +2,7 @@
 from parsers import CustomStates
 
 
-def CustomParser(input):
+def CustomParser(input, input_format=None):
     contents = input.split('\n')
 
     len_flag = []

parsers/iupredparser.py

@@ -2,7 +2,7 @@
 from utils.exceptions import InvalidFormat
 
 
-def IupredParser(input):
+def IupredParser(input, input_format=None):
     contents = input.split('\n')
     output = []
 

parsers/npzparser.py

@@ -0,0 +1,42 @@
+import io
+import base64
+import numpy as np
+from utils.exceptions import InvalidFormat
+from utils import get_unique_distances
+
+
+def parse_array(array):
+    # Bin #0 corresponds with d>20A
+    # Bins #1 ~ #36 correspond with 2A<d<20A in increments of 0.5A
+    contacts = np.sum(array[:, :, 1:13], axis=-1)
+    L = contacts.shape[0]
+    BINS = [np.sum(array[:, :, x:x+4], axis=-1) for x in range(1, 37, 4)]
+    BINS.append(array[:, :, 0].copy())
+    array = np.dstack(BINS)
+    dist_bins = np.nanargmax(array, axis=2)
+    dist_prob = np.amax(array, axis=2)
+    return [[i + 1, j + 1, float(contacts[i, j]), int(dist_bins[i, j]), float(dist_prob[i, j])]
+            for i in range(L) for j in range(i + 5, L)]
+
+
+def NpzParser(input, input_format=None):
+    output = []
+    content_type, content_string = input.split(',')
+    try:
+        decoded = base64.b64decode(content_string)
+        archive = np.load(io.BytesIO(decoded), allow_pickle=True)
+        array = archive['dist']
+        tmp_output = parse_array(array)
+    except (OSError, KeyError, IndexError) as e:
+        raise InvalidFormat('Unable to parse distance NPZ file')
+
+    for contact in tmp_output:
+        # contact = [res_1, res_2, raw_score, distance_bin, distance_score]
+        contact[:2] = sorted(contact[:2], reverse=True)
+        output.append((tuple(contact[:2]), *contact[2:]))
+
+    if not output:
+        raise InvalidFormat('Unable to parse NPZ file')
+    else:
+        unique_contacts = get_unique_distances(output)
+        return unique_contacts

parsers/pdbparser.py

@@ -73,7 +73,7 @@ def remove_atoms(chain):
                 chain[residue.id].detach_child(atom.id)
 
 
-def PDBParser(input):
+def PDBParser(input, input_format=None):
     try:
         parser = BioPDBParser().get_structure('pdb', io.StringIO(input))
         chain = list(parser.get_chains())[0]

parsers/psipredparser.py

@@ -12,7 +12,7 @@ def guess_psipred_format(contents):
     raise InvalidFormat('Unable to guess psipred file format')
 
 
-def PsipredParser(input):
+def PsipredParser(input, input_format=None):
     contents = input.split('\n')
     parser = guess_psipred_format(contents)
     return parser(contents)

parsers/tests/test_distogramparser.py renamed to parsers/tests/test_casprr2parser.py

@@ -1,9 +1,9 @@
 import unittest
-from parsers import DistogramParser
+from parsers import CASPRR2Parser
 from utils.exceptions import InvalidFormat
 
 
-class ContactParserTestCase(unittest.TestCase):
+class CASPRRMODE2ParserTestCase(unittest.TestCase):
 
     def test_1(self):
         dummy_prediction = """PFRMAT RR
@@ -33,7 +33,7 @@ def test_1(self):
         expected_bin_distance = [0, 0, 1, 0, 0, 0, 1, 2, 2, 1, 2, 3]
         expected_bin_score = [0.345, 0.34, 0.56, 0.34, 0.33, 0.33, 0.51, 0.305, 0.3, 0.2, 0.2, 0.3]
 
-        output = DistogramParser(dummy_prediction)
+        output = CASPRR2Parser(dummy_prediction)
 
         self.assertEqual('DISTO', output.pop(-1))
         self.assertEqual(12, len(output))
@@ -49,5 +49,5 @@ def test_2(self):
 100 8 5.382865
 """
         with self.assertRaises(InvalidFormat):
-            output = DistogramParser(dummy_prediction)
+            output = CASPRR2Parser(dummy_prediction)
             self.assertListEqual(output, [])

parsers/topconsparser.py

@@ -2,7 +2,7 @@
 from utils.exceptions import InvalidFormat
 
 
-def TopconsParser(input):
+def TopconsParser(input, input_format=None):
     contents = input.split('\n')
 
     try:

requirements.txt

@@ -19,3 +19,4 @@ visdcc~=0.0.40
 yagmail~=0.14.245
 keyring~=22.0.1
 keyrings.cryptfile~=1.3.6
+numpy~=1.19.4