Scattered data that needs internal reuse
Samples, projects, results, and supplementary tables are scattered across members and storage locations
Initial directionAssess an internal group resource first, starting with the most frequently used data.
GS · 01 — RESEARCH ASSET INDEX
Research data resource consulting and development
Is it worth turning your research into a database now?
If you already have a batch of data, a paper, or a running platform, we first look at what research question it serves, whether existing resources already cover it, how the group will use it, and who maintains it. Once the direction holds up, we discuss whether it should become an internal resource, a public database, or an upgrade of what you already have.
Some projects can move ahead; others are better served by organizing the data first. When the conditions are not there yet, we will say so directly.
See which situation fits your projectFour common directionsNo raw data needed for a first conversation
What we look atGS · v.01
We start with these questions
QUESTION
Research question
DATA
Original data
USAGE
User tasks
PEERS
Existing resources
ACCESS
Access boundaries
GAP
Development needs
Our suggestion
Whether to build a resource · Which direction fits · What to do next
Public project experience · verifiable evidence
These projects have public sites or papers; the scope of our role is stated separately
EDomics
NAR 2023
SilkMeta
NAR 2024
ISwine
Communications Biology 2020
IFmut
JGG 2025
Common starting situations
First, see which situation your project is closest to
Name what is in the way before deciding what to build. This gives an initial direction only; the full assessment lives on its own page.
Original work ready for public sharing
A relatively complete data collection has formed around a species, disease, cohort, or research theme
Initial directionAssess a public thematic data resource first, confirming ownership, licensing, and future update sources.
Multiple projects and omics layers to integrate
Explainable links exist across projects, batches, or omics layers
Initial directionDesign the resource boundary and evidence relationships first, then decide between an internal resource or a research-enhancement track.
An existing platform that is hard to maintain or extend
A database, thematic site, or analysis tool exists, but access, deployment, or updates are unstable
Initial directionStart with an inventory of the legacy platform's assets and data portability, then scope the upgrade.
An analysis tool to bring online
A pipeline, scripts, or reusable results exist and the user tasks are fairly clear
Initial directionAssess putting the tool online or integrating it with a thematic resource; a full database is not the default.
Only a vague idea so far
There is only a rough idea of a website or database; the data boundary is still unclear
Initial directionWe do not recommend building right away; start with a data inventory, direction review, and feasibility assessment.
The first conversation
We start with a clear opinion
It may be to move ahead, to organize data first, to clarify the scope for public release, or not to build for now. The reasoning comes with it.
Whether the work is worth organizing into a resource
We first ask whether it has lasting reuse, search, browsing, or citation value. The answer may be to proceed, to organize data first, or to hold off on development.
Which direction fits better
Considering the primary users, the scope of openness, and maintenance conditions, we judge whether an internal resource, shared collaboration, a public thematic resource, an online tool, or a site upgrade fits best.
What to do next
We point out the gap that most affects the route: data boundaries, ownership and public scope, main use tasks, responsible persons, or maintenance conditions.
This is a preliminary judgment from a first conversation; it cannot decide project approval, procurement, or publication. Public or paper-oriented routes still need a separate check against comparable resources and evidence requirements.
Four common directions
The same data can support very different resources
Some resources serve only the research group; others face peers publicly. Projects with an existing platform usually also weigh repair, migration, or a rebuild.
Internal research-group resource
Fits when
Multiple projects, sample batches, and analysis results have accumulated over years, but the data is scattered and the group needs to keep finding and reusing it.
What it becomes
A searchable, traceable internal data asset that supports project continuity and onboarding.
Clarify first
Needs a named data owner and ongoing use cases; publication is not a prerequisite.
Public thematic data resource
Fits when
A relatively clear body of data around a species, disease, cohort, or research theme, with definable public boundaries.
What it becomes
Browsing, search, download, and citation entry points so public data can be understood and reused by peers.
Clarify first
Ownership, privacy and licensing, and long-term maintenance sources must be confirmed first.
Resource-paper database track
Fits when
Original data, a clear scientific question, and comparable domain resources exist, with a defensible claim of independent contribution and community value.
What it becomes
The data resource, an online database, and the groundwork for a resource paper aligned with the research question.
Clarify first
Whether submission makes sense depends on data scale, differentiation, evidence chain, and maintenance plan; building itself promises no publication.
Upgrading an existing platform
Fits when
A database, thematic site, or analysis tool already exists but is hard to maintain, aging structurally, or needs extension or migration.
What it becomes
Restructuring, extension, migration, or bringing online while keeping the working data and user paths.
Clarify first
Start by confirming the legacy assets, data portability, current access, and the responsibility boundary after the upgrade.
Is now the time to talk about a database
Some projects are ready to move; others need to organize first
As a rule, there should be material worth organizing, a clear idea of who will use it, and someone responsible for openness and maintenance.
Ready to talk now
- You already have data, papers, supplementary material, or public sources worth organizing
- You can describe the study subject, thematic scope, and primary users
- Platform design, development, deployment, or maintenance needs outside help
- Someone knows the data origins and can confirm the public scope and future updates
Better to wait
- There is only the idea of a website; the research task is not settled
- Data origins, ownership, or public scope cannot be stated clearly yet
- No one is currently expected to keep using the resource
- The group already has full platform capability and no current need for outside collaboration
The build phase
Once the direction is set, we build the platform
Each project needs a different depth. Some focus on clarifying data relationships; others on legacy migration and maintenance.
Set the direction and scope
Combining the research question, existing analyses, and future use, decide how deep this project goes and what stays out for now.
Clarify how the data relates
Map samples, projects, analysis results, and origins to one another, so later users know where the data came from and how it can be used.
Build the resource for real
Deliver the database, search, detail pages, visualizations, download, and deployment within the agreed scope — a working entry point, not a mockup.
Spell out how it runs afterward
Before launch, settle access, updates, versioning, and maintenance so the project is not left without an owner.
Judgment examples and project experience
See how we judge, and what we have done
On the left, an example that shows our reasoning. On the right, project experience with public sites or papers to verify, and our role stated separately.
Judgment example · not an actual project
An internal resource fits this data better for now
This example only illustrates how we judge.
- · Several batches of omics results are at hand, but the primary users and public scope are still unclear.
- · We would recommend an internal group resource first, deferring a public database.
- · Once public-ready data, peer use cases, and long-term maintenance ownership are clear, the route can be revisited.
Public project experience · verifiable evidence
Database project experience our core members can publicly describe
Beyond project names, we show where the public evidence is and which parts we actually worked on.
- · Project sites and paper links can be checked publicly.
- · The page lists only the roles GeneSight can publicly describe.
- · Representative projects include EDomics, SilkMeta, ISwine, and IFmut.
Describe what you have, and we will take a first look with you
A first conversation does not require uploading raw data. Just describe the study subject, what you already have, who will use it, and where things are stuck. We will give a preliminary opinion and what still needs confirming.