Standards on SLAP2 (Random Access Projection Microscopy) acquisition¶

Version¶

0.1.0

Introduction¶

This document describes the standards for SLAP2 optical physiology data assets. The standard covers both static SLAP2 structure acquisitions and dynamic SLAP2 experiment acquisitions. Derived SLAP2 data are not yet specified in this standard and are planned for a later version.

Acquisition/Raw/Primary Data Format¶

Following SciComp standards, SLAP2 data MUST be saved in the slap2 modality folder. The timestamped session directory is the data asset root and contains the Allen Institute / Neural Dynamics metadata files for the acquisition. Other modality folders, such as behavior and behavior-videos, MAY exist alongside slap2 and SHOULD follow their own standards.

Within the slap2 folder, a session-specific 1p vasculature image, when taken and retained with the asset, MUST be stored as session_vasculature_1p.tif. If that image is not available with the asset, acquisition.json SHOULD document the imaging location coordinates. vasculature_map_annotated.tif SHOULD be included when available.

File format¶

Static SLAP2 structure assets MUST store their modality-specific files in slap2/static_data. A common naming convention is:

📂slap2
┣ 📜vasculature_map_annotated.tif (preferred)
┣ 📜session_vasculature_1p.tif (if collected)
┗ 📂static_data
  ┣ 📜structure_YYYYMMDD_HHMMSS_DMD#(_CONFIG#).meta
  ┣ 📜structure_YYYYMMDD_HHMMSS_DMD#(_CONFIG#).dat
  ┣ 📜structure_YYYYMMDD_HHMMSS_DMD#(_CONFIG#).tif
  ┗ 📜structure_YYYYMMDD_HHMMSS_DMD#(_CONFIG#)-REFERENCE.tif

Dynamic SLAP2 experiment assets MUST store their modality-specific files in slap2/dynamic_data. A common naming convention is:

📂slap2
┣ 📜vasculature_map_annotated.tif (preferred)
┣ 📜session_vasculature_1p.tif (if collected)
┗ 📂dynamic_data
  ┣ 📜acquisition_YYYYMMDD_HHMMSS_DMD#.meta
  ┣ 📜acquisition_YYYYMMDD_HHMMSS_DMD#-TRIAL######(-CYCLE######).dat
  ┣ 📜acquisition_YYYYMMDD_HHMMSS_DMD#-TRIAL######(-CYCLE######).tif
  ┣ ...
  ┗ 📂reference_stack
    ┣ 📜refStack_YYYYMMDD_HHMMSS_DMD#(_CONFIG#)-REFERENCE.tif
    ┣ 📜refStack_YYYYMMDD_HHMMSS_DMD#(_CONFIG#).meta (optional)
    ┣ 📜refStack_YYYYMMDD_HHMMSS_DMD#(_CONFIG#).dat (optional)
    ┗ 📜refStack_YYYYMMDD_HHMMSS_DMD#(_CONFIG#).tif (optional)

Format specifications¶

The filename stems encode the acquisition time and the DMD index used for the acquisition.

The structure_, acquisition_, and refStack_ prefixes shown above are the preferred short-form conventions for new assets. Alternative filename stems are also accepted.

Beyond that leading stem, the punctuation, datetime format, and suffix ordering shown above are determined by the MBF Bioscience SLAP2 microscope software and are documented here as generated rather than redefined by the Allen Institute / Neural Dynamics. These filenames do not follow the datetime and word-separator conventions in core-standards.md because the naming is generated by the MBF Bioscience SLAP2 software and is outside the control of the Allen Institute / Neural Dynamics.

In every case, companion files MUST still share the same filename stem and preserve the timestamp, DMD, trial, cycle, and -REFERENCE semantics.

Dynamic acquisitions MAY omit the per-trial TIFF files when the SLAP2 acquisition mode does not generate them, but the .dat payloads and their metadata are still required.

The vasculature_map_annotated.tif file is preferred and, when present, is expected to be copied from the mouse-level vasculature reference maintained outside the session asset. session_vasculature_1p.tif captures the session-specific vasculature image used with the SLAP2 acquisition. If no session-specific 1p vasculature image is available with the asset, acquisition.json SHOULD document the imaging location coordinates instead.

Within reference_stack, a -REFERENCE.tif image matching the reference-stack filename stem is required. The corresponding .meta, .dat, and .tif files are optional when the same reference-stack content is already available through a separate static SLAP2 asset.

Application notes¶

Raw data can be read using one of the following libraries:

MATLAB¶

Library with only SLAP2 data reader capabilities: https://github.com/m-xie/Slap2DataReader

Python¶

slap2-utils: https://github.com/Peter-Hogg/SLAP2_Utils

Vocabulary¶

Static data

SLAP2 structure-acquisition files stored under slap2/static_data. These data are meant to be averaged over time and typically capture structural or reference information.

Dynamic data

SLAP2 experiment-acquisition files stored under slap2/dynamic_data. These data are meant to be analyzed in the full spatial (XYZ) and temporal (T) dimensions and typically correspond to the recordings collected during an experiment session.

Reference stack

The reference_stack subdirectory within slap2/dynamic_data. It contains the required reference image (*-REFERENCE.tif) that was used for imaging ROI selection and online motion correction during the dynamic acquisition. If the reference image is not already in its own separate static data asset, the reference_stack subdirectory SHOULD also include the raw data files (*.dat, *.meta).

Imaging regions of interest (ROIs)

SLAP2 is a random access projection microscope, which enables imaging to be targeted only to specified locations. Imaging ROIs are the raster and/or integration ROIs that are selected for imaging on SLAP2.

Field of view (FOV)

Aligned with the Slap2Plane concept in aind-data-schema, a FOV is a specific set of imaging ROIs that are imaged simultaneously. If any imaging ROIs in the set are modified, deleted, or added, the FOV becomes a different FOV.