Glossary

ADAMACS

Architectures for Data Management and Computational Support, the DataJoint-centered pipeline ecosystem used in the lab.

DataJoint schema

Logical table namespace (for example subject, imaging, model, mocap).

Session

A recording unit identified by a unique session_id.

Scan

A recording sub-unit linked to a session, identified by scan_id.

*Task table

Manual/staging table indicating that downstream computed processing should be run.

Worker

Background process that continuously calls populate(..., reserve_jobs=True) on selected tables.

Reserve jobs

DataJoint mechanism to avoid duplicate compute across concurrent workers.

Blob store

Filesystem-backed large-object storage used for large payloads instead of direct DB inline storage.

Curation

Manual/algorithmic quality-control state for extracted imaging components.

paramset_idx

Processing parameter set identifier used in imaging processing workflows.

curation_id

Version key identifying curation iteration.

PyRAT

Animal metadata source system queried for subject/genotype metadata staging.

RSpace

Electronic lab notebook service used optionally for ingest-linked document creation.

CASCADE

Ground-truth-calibrated spike inference integration path in imaging activity workflows.

DLC

DeepLabCut model workflows for video-based pose tracking.

Worker-owned population

Population step that should usually be run by dedicated server worker accounts rather than personal user accounts.

Legacy monorepo

Original adamacs repository before ingest/analysis split.

Split repos

Current recommended working repos:

  • adamacs_ingest

  • adamacs_analysis