DS Administrator Guide
Role Overview
The DS Administrator (Data Science Administrator) is responsible for the day-to-day operational management of projects, pipelines, data access, and compute resources within Quark V3.
The DS Administrator is the role closest to the researchers — ensuring they have the pipelines, workstations, and data access they need to do their work, within the governance boundaries the platform enforces.
Navigation Menu
When logged in as a DS Administrator, the left-hand navigation pane provides access to the following sections:
| Menu Item | Description |
|---|---|
| Dashboard | At-a-glance view of active workspaces and pipelines. Switch projects, filter by user or status, and navigate quickly to any resource. |
| Cost | Monitor budget allocation and resource expenditure across teams, projects, and users. Filter by resource type and date range. |
| HealthOmics Pipelines | Import and configure AWS HealthOmics workflows — both Private and Ready2Run — so researchers can discover and launch them. |
| Project Members | View and search the members of each project, check user status, and see when status was last updated. |
| Apps | Create and manage visualisation applications (e.g., IGV) that researchers use to explore pipeline outputs interactively. |
| Datasets | Browse the dataset catalog, review aggregate statistics, and monitor cohort requests and approval timelines. |
| Workstations | Monitor the status of user workstations, provision new workstations from templates, and connect to or launch existing instances. |
| Metadata | Configure global and pipeline-level metadata validation rules to enforce consistent, discoverable data standards across the platform. |
| Requests | Review and action user requests for data downloads, data uploads, workstation access, and cohort approvals. |
| Audit Logs | Access the immutable record of all platform activity — filter by user, resource type, and date range to support compliance and investigation. |
Core Responsibilities
Pipeline Enablement
The DS Administrator imports available HealthOmics pipelines. Importing pipelines, setting meaningful parameter defaults, and configuring metadata validation for each pipeline are foundational tasks that determine whether researchers can self-serve effectively.
Data Governance Oversight
The DS Administrator monitors what data is being accessed, by whom, and for what purpose. This involves reviewing cohort requests in Datasets, actioning access requests in Requests, and cross-referencing activity in Audit Logs.
Compute and Cost Management
The DS Administrator tracks resource expenditure in Cost and manages the workstations that researchers use to conduct their analyses. For example, ensuring workstations are appropriately provisioned, not left running unnecessarily, and within budget.
Metadata Standards Enforcement
Consistent, validated metadata is what makes data discoverable and reusable across pipelines and projects. The DS Administrator configures and maintains the global and pipeline-level metadata rules in Metadata that ensure researchers can find, filter, and compare data reliably.
Suggested Starting Point
- Confirm project membership is correct — Project Members
- Import the pipelines the team needs — HealthOmics Pipelines
- Configure metadata validation for those pipelines — Metadata
- Provision workstations from appropriate templates — Workstations
- Review any pending access requests — Requests
- Set a regular cadence for reviewing Cost and Audit Logs — Cost | Audit Logs