Merge pull request #134 from rigdenlab/development

Minor update

Author: FilomenoSanchez

.github/workflows/main.yml

@@ -39,6 +39,7 @@ jobs:
           redis-version: 5
       - name: Run tests.py
         env:
+          THIS_IS_GH_ACTIONS: 1
           KEYDB_URL: $ {{ secrets.KEYDB_URL }}
         run: |
           python tests.py

CHANGELOG.rst

@@ -2,6 +2,20 @@ Changelog
 =========
 
 
+0.4.1
+-----
+
+Added
+~~~~~
+- Added contact diff track with smoothed values (MCC for contacts and RMSD for distograms)
+
+Changed
+~~~~~
+- Increased contrast in sequence hydrophobicity color palettes
+- Use AMISE to estimate bandwidth required to calculate contact density
+- Updated track selector layout
+
+
 0.4
 -----
 

README.md

@@ -59,21 +59,25 @@ Once you have installed `redis`, you will need to start the service by running:
 $   sudo service redis start
 ```
 
-You will also need to create a environment variable called `KEYDB_URL` with 
-the URL to connect to the redis server you just started on your machine:
+Now you'll need to clone this repository, install the requirements and setup environment variables. 
+Please note that ConPlot requires at least `python 3.6`.
 
 ```bash
-$   KEYDB_URL=redis://localhost:6379
+$   git clone https://github.com/rigdenlab/conplot
+$   cd conplot
+$   python3.6 -m pip install -r requirements.txt
+$   echo "KEYDB_URL=0://localhost:6379" > .env
+$   echo "KEYDB_TIMEOUT=3600" >> .env 
 ```
 
-After this, all you need to do is clone this repository, install the requirements 
-and start the Flask development server on your machine. Please note that ConPlot 
-requires at least `python 3.6`.
+With the last two commands you will also have created an `.env` file with a variable named 
+`KEYDB_URL` pointing to the redis server and a `KEYDB_TIMEOUT` variable with the session 
+timeout value. This is the time at which a session expires after inactivity. By default in 
+`www.conplot.org` this has a value of 3600 minutes, but if running locally you can set this 
+time to any other value. The only thing left to do is to start the Flask development 
+server on your machine:
 
 ```bash
-$   git clone https://github.com/rigdenlab/conplot
-$   cd conplot
-$   python3.6 -m pip install -r requirements.txt
 $   python3.6 app.py
 ```
 

app.py

@@ -24,7 +24,7 @@ def serve_layout():
     except (keydb.ConnectionError, TypeError, KeyError) as e:
         app.logger.error('Redis connection error! {}'.format(e))
         return layouts.RedisConnectionError()
-    session_id = session_utils.initiate_session(cache, app.logger)
+    session_id = session_utils.initiate_session(cache, app.logger, keydb_timeout)
     return layouts.Base(session_id)
 
 
@@ -44,6 +44,7 @@ def serve_layout():
         'requests_pathname_prefix': '/conplot/',
     })
 keydb_pool = keydb_utils.create_pool(os.environ.get('KEYDB_URL'))
+keydb_timeout = os.environ.get('KEYDB_TIMEOUT')
 app.layout = serve_layout
 
 
@@ -407,7 +408,7 @@ def javascript_exe_button(n_clicks, session_id):
 
     elif 'new-session' in trigger['prop_id'] or session_utils.is_expired_session(session_id, cache, app.logger):
         cache = keydb.KeyDB(connection_pool=keydb_pool)
-        new_session_id = session_utils.initiate_session(cache, app.logger)
+        new_session_id = session_utils.initiate_session(cache, app.logger, keydb_timeout)
         return "location.reload();", no_update, new_session_id
 
     else:
@@ -454,7 +455,7 @@ def create_ConPlot(plot_click, refresh_click, factor, contact_marker_size, track
     if any([True for x in (factor, contact_marker_size, track_marker_size, track_separation) if x is None or x < 0]):
         app.logger.info('Session {} invalid display control value detected'.format(session_id))
         return no_update, components.InvalidInputModal(), no_update, no_update
-    elif superimpose and ('---' in cmap_selection or len(set(cmap_selection)) == 1):
+    elif superimpose and ('--- Empty ---' in cmap_selection or len(set(cmap_selection)) == 1):
         return no_update, components.InvalidMapSelectionModal(), no_update, no_update
 
     app.logger.info('Session {} creating conplot'.format(session_id))

components/__init__.py

@@ -11,6 +11,7 @@ class UserReadableTrackNames(Enum):
     heatmap = 'Heatmap'
     hydrophobicity = 'Hydrophobicity'
     density = 'Contact density'
+    diff = 'Contact diff'
 
 
 class EmailIssueReference(Enum):

components/cards.py

@@ -1,3 +1,4 @@
+from utils import cache_utils
 import components
 import dash_core_components as dcc
 import dash_bootstrap_components as dbc
@@ -286,45 +287,47 @@ def DisplayControlCard(available_tracks=None, selected_tracks=None, selected_cma
                             html.H5("Active tracks", className="card-text", style={'text-align': "center"}),
                             html.Hr(),
                             html.Br(),
-                            TrackSelectionCard('-4', selected_tracks[0], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('-4', available_tracks, selected_tracks[0]), outline=False),
                             html.Br(),
-                            TrackSelectionCard('-3', selected_tracks[1], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('-3', available_tracks, selected_tracks[1]), outline=False),
                             html.Br(),
-                            TrackSelectionCard('-2', selected_tracks[2], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('-2', available_tracks, selected_tracks[2]), outline=False),
                             html.Br(),
-                            TrackSelectionCard('-1', selected_tracks[3], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('-1', available_tracks, selected_tracks[3]), outline=False),
                             html.Br(),
-                            TrackSelectionCard(' 0', selected_tracks[4], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('0', available_tracks, selected_tracks[4]), outline=False),
                             html.Br(),
-                            TrackSelectionCard('+1', selected_tracks[5], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('+1', available_tracks, selected_tracks[5]), outline=False),
                             html.Br(),
-                            TrackSelectionCard('+2', selected_tracks[6], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('+2', available_tracks, selected_tracks[6]), outline=False),
                             html.Br(),
-                            TrackSelectionCard('+3', selected_tracks[7], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('+3', available_tracks, selected_tracks[7]), outline=False),
                             html.Br(),
-                            TrackSelectionCard('+4', selected_tracks[8], available_tracks=available_tracks),
+                            dbc.Card(components.TrackLayoutSelector('+4', available_tracks, selected_tracks[8]), outline=False),
                             html.Br(),
                             html.Br(),
                             html.H5("Colour palettes", className="card-text", style={'text-align': "center"}),
                             html.Hr(),
                             html.Br(),
                             ColorPaletteSelectionCard('density', selected_palettes[0]),
                             html.Br(),
-                            ColorPaletteSelectionCard('custom', selected_palettes[1]),
+                            ColorPaletteSelectionCard('diff', selected_palettes[1]),
                             html.Br(),
-                            ColorPaletteSelectionCard('heatmap', selected_palettes[2]),
+                            ColorPaletteSelectionCard('custom', selected_palettes[2]),
                             html.Br(),
-                            ColorPaletteSelectionCard('hydrophobicity', selected_palettes[3]),
+                            ColorPaletteSelectionCard('heatmap', selected_palettes[3]),
                             html.Br(),
-                            ColorPaletteSelectionCard('membranetopology', selected_palettes[4]),
+                            ColorPaletteSelectionCard('hydrophobicity', selected_palettes[4]),
                             html.Br(),
-                            ColorPaletteSelectionCard('msa', selected_palettes[5]),
+                            ColorPaletteSelectionCard('membranetopology', selected_palettes[5]),
                             html.Br(),
-                            ColorPaletteSelectionCard('conservation', selected_palettes[6]),
+                            ColorPaletteSelectionCard('msa', selected_palettes[6]),
                             html.Br(),
-                            ColorPaletteSelectionCard('disorder', selected_palettes[7]),
+                            ColorPaletteSelectionCard('conservation', selected_palettes[7]),
                             html.Br(),
-                            ColorPaletteSelectionCard('secondarystructure', selected_palettes[8]),
+                            ColorPaletteSelectionCard('disorder', selected_palettes[8]),
+                            html.Br(),
+                            ColorPaletteSelectionCard('secondarystructure', selected_palettes[9]),
                             html.Br(),
                         ])
                     ]
@@ -334,14 +337,6 @@ def DisplayControlCard(available_tracks=None, selected_tracks=None, selected_cma
     else:
         raise ValueError('This should not occur! Please report.')
 
-
-def TrackSelectionCard(track_idx, track_value, available_tracks):
-    track_options = [{'label': '---', 'value': '---'}]
-    track_options += [{'label': fname, 'value': fname} for fname in available_tracks]
-
-    return dbc.Card(components.TrackLayoutSelector(track_idx, track_options, track_value), outline=False)
-
-
 def ColorPaletteSelectionCard(dataset, selected_palette):
     available_palettes = []
     for palette in color_palettes.DatasetColorPalettes.__getattr__(dataset).value:
@@ -351,7 +346,7 @@ def ColorPaletteSelectionCard(dataset, selected_palette):
 
 
 def HalfSquareSelectionCard(square_idx, selection, available_cmaps):
-    cmap_options = [{'label': '---', 'value': '---'}]
+    cmap_options = [{'label': '--- Empty ---', 'value': '--- Empty ---'}]
     cmap_options += [{'label': fname, 'value': fname} for fname in available_cmaps]
 
     return dbc.Card(components.HalfSquareSelector(square_idx, cmap_options, selection), outline=False)

components/listgrpoups.py

@@ -191,48 +191,38 @@ def AdjustPlotHelpList():
                  'series of input menus:',
                  html.Ul([
                      html.Li(['L/N selector: Change the values of ', html.I('N'),
-                              ' with this selector to choose how many contacts should be '
-                              'included in the plot (L is the number of residues in the '
-                              'protein sequence, residues are sorted by their probability '
-                              'score). If you set ', html.I('N'),
-                              ' to 0, then all contacts in the file will be displayed. Please '
-                              'note that only numerical values between 0 and 10 are recommended.']),
-                     html.Li('Size selector: Change the size of the contact markers in the '
-                             'plot. ConPlot will set a default value depending on the size of '
-                             'the protein you are working with, but you can still change this '
-                             'if you would like to make the markers smaller or bigger. Please '
+                              ' with this selector to choose how many contacts should be included in the plot (L is '
+                              'the number of residues in the  protein sequence, residues are sorted by their '
+                              'probability score). If you set ', html.I('N'),
+                              ' to 0, then all contacts in the file will be displayed. Please note that only numerical '
+                              'values between 0 and 10 are recommended. Additionally, please remember that contact '
+                              'data shown for PDB files is unaltered by this selector.']),
+                     html.Li('Size selector: Change the size of the contact markers in the plot. ConPlot will set a '
+                             'default value depending on the size of the protein you are working with, but you can '
+                             'still change this if you would like to make the markers smaller or bigger. Please '
                              'note that only numerical values between 1 and 15 are recommended.'),
-                     html.Li(['Map A and Map B selectors: These two selectors let you choose '
-                              'which contact data should be displayed on the plot. By '
-                              'default, ', html.I('Map A'),
-                              ' refers to the top half triangle of the map, and ',
-                              html.I('Map B'), ' to the lower one. If the ',
-                              html.I('Superimpose Maps'),
-                              ' switch is activated, then the roles of these two dropdown '
-                              'menus change: ', html.I('Map A'),
-                              ' is now used to select the reference map, which will be '
-                              'compared with the secondary map selected with the ',
-                              html.I('Map B'), ' selector.']),
-                     html.Li(['Superimpose Maps Switch: As explained above, if this switch '
-                              'is activated ', html.I('Map A'),
-                              ' will be used as a reference map to be compared with ',
-                              html.I('Map B'),
-                              '. In this mode, contacts will be coloured according to their '
-                              'presence in the reference map and the secondary map. Contacts '
-                              'that appear on both the reference and the secondary map will be '
-                              'coloured in black -match-, those that only appear in the '
-                              'reference in grey -absent-, and those that only appear in the '
-                              'secondary map in red -mismatch-. Please note that you can only '
-                              'use this mode if you select two different contact map files in ',
+                     html.Li(['Map A and Map B selectors: These two selectors let you choose which contact data should '
+                              'be displayed on the plot. By ' 'default, ', html.I('Map A'),
+                              ' refers to the top half triangle of the map, and ', html.I('Map B'),
+                              ' to the lower one. If the ', html.I('Superimpose Maps'),
+                              ' switch is activated, then these roles change: ',  html.I('Map A'),
+                              ' is now used to select the reference map, which will be compared with the secondary map '
+                              'selected with the ', html.I('Map B'), ' selector.']),
+                     html.Li(['Superimpose Maps Switch: As explained above, if this switch is activated ',
+                              html.I('Map A'), ' will be used as a reference map to be compared with ', html.I('Map B'),
+                              '. In this mode, contacts will be coloured according to their presence in the reference '
+                              'map and the secondary map. Contacts that appear on both the reference and the secondary '
+                              'map will be coloured in black -match-, those that only appear in the reference in grey '
+                              '-absent-, and those that only appear in the secondary map in red -mismatch-. Please '
+                              'note that you can only use this mode if you select two different contact map files in ',
                               html.I('Map A'), ' and ', html.I('Map B'), ' selectors.']),
                      html.Li(['Create Heatmap Switch: If this switch is activated, a heatmap will be created with the '
                               'provided residue contact information. By default, if a contact map is uploaded, the '
                               'intensity of the colours in this heatmap will correspond with the confidence of each '
-                              'contact. Alternatively, if a residue-residue distance prediction file has been uploaded '
-                              '(', html.I('CASPRR_MODE2'),
-                              ' format), the heatmap will correspond with the predicted distances for '
-                              'each residue pair oin this file. Please note that when this mode is active, the ',
-                              html.I('L/N'), ' selector and the ', html.I('Size'),
+                              'contact. Alternatively, if a residue-residue distance prediction file has been '
+                              'uploaded, the heatmap will correspond with the predicted distances for each residue '
+                              'pair oin this file. Please note that when this mode is active, the ', html.I('L/N'),
+                              ' selector and the ', html.I('Size'),
                               ' selector will be disabled. You can read more about how to visualise residue-residue '
                               'distance predictions at ',
                               html.I('Tutorial 4. Residue-Residue distance predictions'), '.']),
@@ -242,9 +232,8 @@ def AdjustPlotHelpList():
                              'would normally be displayed.')
                  ])],
                 style={"font-size": "110%", 'text-align': "justify"}),
-        html.Li(['Section 2: Adjust additional tracks. In this section you will find selectors '
-                 'that will let you control aspects about how the additional tracks are being '
-                 'displayed in the plot:',
+        html.Li(['Section 2: Adjust additional tracks. In this section you will find selectors that will let you '
+                 'control aspects about how the additional tracks are being displayed in the plot:',
                  html.Ul([
                      html.Li('Size selector: Change the size of the tiles used to create the '
                              'tracks on the diagonal of the plot. By changing this value, '
@@ -326,6 +315,15 @@ def AdditionalFormatsHelpList():
                  html.A(html.U('here'), href=UrlIndex.CONSURF_CITATION.value),
                  '.'],
                 style={"font-size": "110%", 'text-align': "justify"}),
+        html.Li(['A3M file. This is a multiple sequence alignment file that should have been obtained using the '
+                 'sequence of interest as a query. ConPlot will parse the file and calculate the MSA coverage along '
+                 'the query sequence, normalise these values (1-10) and create a track where each residue '
+                 'is coloured according to the number of sequences aligned in that particular position These '
+                 'files are used in most contact prediction pipelines, and visualising the MSA coverage can help you '
+                 'understand the quality of the information used to obtain your predictions. Several alignment tools '
+                 'will create MSA files in this format, like for example HHBLITS, which you can use '
+                 'online ', html.A(html.U('here'), href=UrlIndex.HHBLITS_URL.value), '.'],
+                style={"font-size": "110%", 'text-align': "justify"}),
         html.Li(['CUSTOM file. These files are plain text files that can be created manually '
                  'by users to include additional tracks of information to the plot. These '
                  'files enable limitless personalisation of the contact map plot, as it '