Merge branch 'release/2.29.0'

Author: oliche

.github/workflows/ibllib_ci.yml

@@ -10,9 +10,28 @@ on:
     branches: [ master, develop ]
 
 jobs:
+  detect-outstanding-prs:  # Don't run builds for push events if associated with PR
+    runs-on: ubuntu-latest
+    env:
+      GH_TOKEN: ${{ github.token }}
+    outputs:
+      abort: ${{ steps.debounce.outputs.abort }}
+    steps:
+      - name: Debounce
+        if: github.event_name == 'push'
+        id: debounce
+        run: |
+          pr_branches=$(gh pr list --json headRefName --repo $GITHUB_REPOSITORY)
+          if [[ $(echo "$pr_branches" | jq -r --arg GITHUB_REF '.[].headRefName | select(. == $GITHUB_REF)') ]]; then
+            echo "This push is associated with a pull request. Skipping the job."
+            echo "abort=true" >> "$GITHUB_OUTPUT"
+          fi
+
   build:
     name: build (${{ matrix.python-version }}, ${{ matrix.os }})
     runs-on: ${{ matrix.os }}
+    needs: debounce
+    if: needs.debounce.outputs.abort != 'true'
     strategy:
       fail-fast: false  # Whether to stop execution of other instances
       max-parallel: 2

brainbox/io/one.py

@@ -28,6 +28,7 @@
 from ibllib.plots import vertical_lines
 
 import brainbox.plot
+from brainbox.io.spikeglx import Streamer
 from brainbox.ephys_plots import plot_brain_regions
 from brainbox.metrics.single_units import quick_unit_metrics
 from brainbox.behavior.wheel import interpolate_position, velocity_filtered
@@ -793,6 +794,7 @@ class SpikeSortingLoader:
     datasets: list = None   # list of all datasets belonging to the session
     # the following properties are the outcome of a reading function
     files: dict = None
+    raw_data_files: list = None  # list of raw ap and lf files corresponding to the recording
     collection: str = ''
     histology: str = ''  # 'alf', 'resolved', 'aligned' or 'traced'
     spike_sorter: str = 'pykilosort'
@@ -829,6 +831,7 @@ def __post_init__(self):
         if self.atlas is None:
             self.atlas = AllenAtlas()
         self.files = {}
+        self.raw_data_files = []
 
     def _load_object(self, *args, **kwargs):
         """
@@ -881,6 +884,11 @@ def load_spike_sorting_object(self, obj, *args, **kwargs):
         self.download_spike_sorting_object(obj, *args, **kwargs)
         return self._load_object(self.files[obj])
 
+    def get_version(self, spike_sorter='pykilosort'):
+        collection = self._get_spike_sorting_collection(spike_sorter=spike_sorter)
+        dset = self.one.alyx.rest('datasets', 'list', session=self.eid, collection=collection, name='spikes.times.npy')
+        return dset[0]['version'] if len(dset) else 'unknown'
+
     def download_spike_sorting_object(self, obj, spike_sorter='pykilosort', dataset_types=None, collection=None,
                                       missing='raise', **kwargs):
         """
@@ -919,6 +927,46 @@ def download_spike_sorting(self, **kwargs):
             self.download_spike_sorting_object(obj=obj, **kwargs)
         self.spike_sorting_path = self.files['spikes'][0].parent
 
+    def download_raw_electrophysiology(self, band='ap'):
+        """
+        Downloads raw electrophysiology data files on local disk.
+        :param band: "ap" (default) or "lf" for LFP band
+        :return: list of raw data files full paths (ch, meta and cbin files)
+        """
+        raw_data_files = []
+        for suffix in [f'*.{band}.ch', f'*.{band}.meta', f'*.{band}.cbin']:
+            try:
+                # FIXME: this will fail if multiple LFP segments are found
+                raw_data_files.append(self.one.load_dataset(
+                    self.eid,
+                    download_only=True,
+                    collection=f'raw_ephys_data/{self.pname}',
+                    dataset=suffix,
+                    check_hash=False,
+                ))
+            except ALFObjectNotFound:
+                _logger.debug(f"{self.session_path} can't locate raw data collection raw_ephys_data/{self.pname}, file {suffix}")
+        self.raw_data_files = list(set(self.raw_data_files + raw_data_files))
+        return raw_data_files
+
+    def raw_electrophysiology(self, stream=True, band='ap', **kwargs):
+        """
+        Returns a reader for the raw electrophysiology data
+        By default it is a streamer object, but if stream is False, it will return a spikeglx.Reader after having
+        downloaded the raw data file if necessary
+        :param stream:
+        :param band:
+        :param kwargs:
+        :return:
+        """
+        if stream:
+            return Streamer(pid=self.pid, one=self.one, typ=band, **kwargs)
+        else:
+            raw_data_files = self.download_raw_electrophysiology(band=band)
+            cbin_file = next(filter(lambda f: f.name.endswith(f'.{band}.cbin'), raw_data_files), None)
+            if cbin_file is not None:
+                return spikeglx.Reader(cbin_file)
+
     def load_channels(self, **kwargs):
         """
         Loads channels
@@ -1282,7 +1330,8 @@ def load_trials(self):
         """
         # itiDuration frequently has a mismatched dimension, and we don't need it, exclude using regex
         self.one.wildcards = False
-        self.trials = self.one.load_object(self.eid, 'trials', collection='alf', attribute=r'(?!itiDuration).*').to_df()
+        self.trials = self.one.load_object(
+            self.eid, 'trials', collection='alf', attribute=r'(?!itiDuration).*').to_df()
         self.one.wildcards = True
         self.data_info.loc[self.data_info['name'] == 'trials', 'is_loaded'] = True
 

ibllib/__init__.py

@@ -2,7 +2,8 @@
 import logging
 import warnings
 
-__version__ = '2.28.2'
+__version__ = '2.29.0'
+
 warnings.filterwarnings('always', category=DeprecationWarning, module='ibllib')
 
 # if this becomes a full-blown library we should let the logging configuration to the discretion of the dev

ibllib/atlas/genes.py

@@ -1,6 +1,6 @@
 """Gene expression maps."""
 
-from iblatlas import genes
+from iblatlas.genomics import genes
 from ibllib.atlas import deprecated_decorator
 
 

ibllib/io/session_params.py

@@ -318,7 +318,10 @@ def get_task_protocol(sess_params, task_collection=None):
     """
     collections = get_collections({'tasks': sess_params.get('tasks')})
     if task_collection is None:
-        return set(collections.keys())  # Return all protocols
+        if len(collections) == 0:
+            return None
+        else:
+            return set(collections.keys())  # Return all protocols
     else:
         return next((k for k, v in collections.items() if v == task_collection), None)
 

release_notes.md

@@ -1,3 +1,8 @@
+## Release Notes 2.29
+
+### features
+- Added raw data loaders and synchronisation tools in brainbox.io.one.SpikeSortingLoader, method `ssl.raw_electrophysiology()`
+
 ## Release Notes 2.28
 
 ### features

requirements.txt

@@ -5,13 +5,11 @@ flake8>=3.7.8
 globus-sdk
 graphviz
 matplotlib>=3.0.3
-mtscomp>=1.0.1
 numba>=0.56
 numpy>=1.18
 nptdms
 opencv-python-headless
 pandas
-phylib>=2.4
 pyarrow
 pynrrd>=0.4.0
 pytest
@@ -23,11 +21,13 @@ sparse
 seaborn>=0.9.0
 tqdm>=4.32.1
 # ibl libraries
+iblatlas>=0.4.0
 ibl-neuropixel>=0.8.1
 iblutil>=1.7.0
 labcams  # widefield extractor
-ONE-api>=2.5
+mtscomp>=1.0.1
+ONE-api>=2.6
+phylib>=2.4
+psychofit
 slidingRP>=1.0.0  # steinmetz lab refractory period metrics
 wfield==0.3.7  # widefield extractor frozen for now (2023/07/15) until Joao fixes latest version
-psychofit
-iblatlas