DEV Community

Cover image for Your First Mini-Curation: Getting Started and Lessons from My Experience

Your First Mini-Curation: Getting Started and Lessons from My Experience

In my previous article, Understanding BugSigDB and Its Curation Policy, I introduced BugSigDB, explained what microbial signatures are, and discussed why the project needs a consistent approach to curating scientific findings.

Now comes the interesting part: putting that knowledge into practice.

If you're new to BugSigDB, your first mini-curation is an opportunity to move beyond reading documentation and actually work with a scientific paper. You'll begin identifying the information that matters, interpreting microbial findings, and learning how those findings are represented in the database.

I remember approaching my first mini-curation with curiosity, but also with a lot of questions. I had a background in Data Science and experience reading scientific literature from my master's research, but microbiome research and microbial signature curation were new territory.

Here is how I approached it and what I learned along the way.

What Is a Mini-Curation?

A mini-curation is a small introductory curation exercise designed to help new contributors become familiar with the BugSigDB curation process.

Rather than immediately tackling a complex research paper with multiple experiments and comparisons, you get to work through a smaller exercise and understand the basic concepts involved.

You begin by examining a scientific study, identifying the relevant information, interpreting its findings, and representing the microbial signatures according to BugSigDB's Curation Policy.

Think of it as your first opportunity to connect what you've learned from the documentation with the actual work of a curator.

It is not simply about entering information into a database. You need to understand what the researchers investigated, how they conducted their study, which groups they compared, and what their results mean.

Before You Begin

Before starting, make sure you have a few things in place:

  1. Your BugSigDB account: You need access to the database to carry out the curation.
  2. The study or exercise assigned to you: Familiarize yourself with the paper you will be working on.
  3. The Curation Policy: Keep it accessible. You will refer to it frequently.
  4. The project documentation: Revisit the relevant guides whenever you need clarification.

One thing I would emphasize is that you don't need to memorize the entire Curation Policy before beginning. You will understand many of its instructions better when you encounter them in an actual study.

However, you should know where to find the information and be willing to consult it.

Working Through Your First Mini-Curation

Let's look at the main things you need to pay attention to when working through a scientific paper.

1. Understand the Study

Before extracting anything, read the paper to understand what the researchers were investigating.

Start with the title and abstract, then move to the Introduction, Methods, and Results sections.

Ask yourself:

  • What is the research question?
  • What population or subjects were studied?
  • What biological samples were collected?
  • What condition or characteristic is being investigated?
  • What groups are being compared?

The abstract is a useful starting point, but it rarely contains everything you need for a complete curation.

You will often find important details in the Methods section, tables, figures, and supplementary materials.

2. Identify the Study Design

Pay attention to how the researchers conducted the study.

Was it a case-control study? A cross-sectional study? A prospective cohort? Was there an intervention?

The study design helps describe the structure of the research and how participants or subjects were selected and followed.

Don't select a study design simply because the paper mentions a clinical trial or uses the word "cohort." Read the methodology and determine which design best represents the actual study.

If you are uncertain, revisit the Curation Policy and ask the community for clarification.

3. Understand the Experimental Comparison

This is one of the most important parts of the process.

You need to identify the groups being compared in the analysis that produced the microbial findings.

For example, a study might compare:

  • Individuals with a particular disease against healthy controls.
  • Participants receiving a treatment against those who did not.
  • Male participants against female participants.
  • Different age groups.

Be precise about the comparison. A paper can contain several analyses, and not every comparison necessarily represents a separate experiment for BugSigDB purposes.

You need to understand which groups are being compared and which findings belong to that comparison.

4. Identify the Relevant Metadata

As you read the paper, look for information about the study subjects and how the samples were collected and analyzed.

Some of the details you may need include:

  • Host species.
  • Country or geographical location.
  • Body site or sample type.
  • Condition being investigated.
  • Sequencing method and platform.
  • Statistical analysis and significance threshold.

The Methods section is particularly useful here.

For example, microbiome studies may use 16S rRNA gene sequencing or whole-metagenome shotgun sequencing. Some papers also specify the variable region sequenced, the sequencing platform, and the statistical methods used to identify differences in microbial abundance.

Don't assume that every detail will be available. Record what the paper actually reports and follow the policy when information is missing.

5. Identify the Microbial Signatures

Now comes the part that requires careful attention to the Results section.

Look for the microbial taxa reported as significantly increased or decreased in the comparison.

You may find these results in paragraphs, tables, figures, or supplementary datasets.

Pay close attention to the direction of change.

For example, suppose researchers compare people living with HIV against people without HIV. If a particular bacterial genus is reported as less abundant in the HIV group, you must represent that finding in the correct direction.

It would be incorrect to record it as increased in the HIV group.

Also, distinguish between findings that meet the study's statistical criteria and those that are merely mentioned in the discussion.

Your job is to curate the findings that meet the relevant requirements of the Curation Policy, not every microbial name appearing in the paper.

6. Enter the Information into BugSigDB

Once you have identified the relevant information, you can begin entering it into the database.

Follow the available fields and instructions carefully. Use the appropriate ontology terms and taxonomic identifiers where required.

You may encounter unfamiliar terminology or find that the wording in the paper does not exactly match the available options in BugSigDB.

This is normal.

Consult the documentation, check the relevant ontology or taxonomy resource, and ask for help when necessary. Avoid selecting a term simply because it looks similar to the one in the paper.

Remember that the purpose of curation is to represent the scientific findings accurately and consistently.

What I Learned from My First Mini-Curation

My first mini-curation involved comparing the gut microbiome of individuals living with HIV against individuals without HIV.

The study examined microbial differences between the groups, and my task involved identifying the relevant microbial findings and representing them correctly in BugSigDB.

It was a relatively small exercise, but it introduced me to several things I would encounter repeatedly in subsequent curations.

Here are some of my biggest takeaways.

1. Reading a Scientific Paper for Curation Is Different

I had read scientific papers before, particularly during my master's research. However, reading a paper to understand its general findings is different from reading one to extract structured information.

As a curator, you are constantly asking specific questions.

Where did the researchers collect their samples? What groups did they compare? Which statistical method did they use? Where exactly are the reported microbial findings?

You begin moving between the Methods, Results, tables, figures, and supplementary materials.

Sometimes, the information you need is not in the section where you initially expect to find it.

I learned that effective curation requires more than reading a paper from beginning to end. It requires knowing what information to look for and where to find it.

2. Direction of Abundance Matters More Than You Think

This was one of the most important lessons from my first exercise.

When recording a microbial signature, you must pay attention to which group has the increased or decreased abundance.

For example, if a bacterium is less abundant in the HIV group, you cannot record it as increased in that group.

That may sound obvious, but when you are working through several findings, it is easy to lose track of the comparison direction.

A small mistake can completely misrepresent what the researchers reported.

My advice is to keep the comparison groups visible while extracting microbial findings. Always ask yourself: increased or decreased in which group?

3. Scientific Terminology and Ontology Terms Are Not Always the Same

Another thing I encountered was the challenge of selecting the appropriate ontology term.

Scientific papers use terminology that may not correspond exactly to the available ontology labels. Some terms may also be outdated, deprecated, or replaced by newer identifiers.

In my mini-curation, I encountered an HIV-related term that was obsolete in the ontology. The replacement term was not immediately obvious to me.

This taught me to look beyond the wording in the paper and check the relevant ontology carefully.

I also learned that it is perfectly reasonable to ask the community when you are unsure. It is much better to clarify a term than to make an unsupported choice.

4. The Curation Policy Is a Working Reference, Not Just Reading Material

Before beginning, I thought of the Curation Policy as documentation I needed to understand.

After working through the mini-curation, I realized that it is something I would continually return to.

You may understand a particular instruction while reading the policy, but applying it to an actual paper can raise new questions.

How should you represent this comparison? Which study design fits best? Does this finding meet the criteria for inclusion?

These are the kinds of questions that make the policy useful.

Don't be discouraged if you find yourself revisiting the same sections. That is part of learning the process.

5. Asking Questions Is Part of Contributing

One of the things I appreciated about the BugSigDB community was the opportunity to ask questions and receive guidance.

When you are new to a project, it is easy to feel that you should figure everything out independently before asking for help.

But open-source contribution is ✨collaborative✨.

Make an effort to investigate the problem first. Read the documentation, examine the paper, and explain what you have tried. Then ask a specific question.

You will not only get help with your immediate task; you will also learn how experienced contributors approach similar problems.

6. Accuracy Is More Important Than Speed

Finally, I learned not to rush.

When you see other contributors making progress, it can be tempting to focus on completing your curation as quickly as possible.

However, scientific curation requires careful interpretation. A wrong comparison, an incorrect abundance direction, or an unsuitable ontology term can affect the quality of the resulting data.

Take your time. Verify your findings. Revisit the paper when something does not make sense.

The goal is not simply to complete the exercise. It is to understand what you are doing.

A Few Tips for Your First Mini-Curation

If I were starting again, these are the things I would remind myself:

  1. Read the paper beyond the abstract.
  2. Keep the Curation Policy open while working.
  3. Pay particular attention to the Methods and Results sections.
  4. Use tables, figures, and supplementary materials to verify findings.
  5. Keep track of the comparison groups when recording abundance directions.
  6. Check ontology terms instead of relying on assumptions.
  7. Write down questions as they arise.
  8. Ask the community when you cannot resolve something confidently.

You don't have to understand everything immediately. The more studies you work through, the more familiar the process becomes.

What's Next?

Completing your first mini-curation is an important step. You have moved from learning about BugSigDB to actually participating in the curation process.

But this is only the beginning.

Working on full studies introduces additional challenges. A paper may contain multiple experiments, several comparisons, different sample types, and microbial findings scattered across numerous tables and supplementary files.

As you gain experience, you will begin to recognize patterns, interpret research methods more confidently, and make better curation decisions.

In the next article, I'll share my transition from the mini-curation to working on full BugSigDB studies, including some of the lessons I learned from my early contributions.

Happy curating!

Top comments (0)