Why Science? | MGnify Microbiome Analysis and Study Evidence

Start with the record itself, because a good science explanation names the object before interpreting it. Microbiome research studies microorganisms associated with an environment or host. MGnify supports several sequence-data approaches, including amplicon, metagenomic, metatranscriptomic and assembly-based analysis, which answer overlapping but different questions. For secondary science learning, that distinction turns database browsing into evidence practice rather than fact collecting.

Try this: Make four cards for amplicon, metagenome, metatranscriptome and assembly; give each one a cautious question it could help investigate. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Do not fill a blank with a guess. Record the missing field, follow the linked accession or method, and state exactly what remains uncertain. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Next chapter

Here the useful move is to separate what the archive stores from what a researcher may later infer. MGnify study records gather context and relationships for a project or dataset. A study is the broad container, while its samples, runs, assemblies and analyses remain distinct objects with their own identifiers. It also connects classroom habits—defining variables, tracing methods and comparing like with like—to real research infrastructure.

Try this: Open a study record and draw its first two layers without counting a run, sample and analysis as equivalent units. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Keep observation and interpretation in separate sentences. That simple discipline makes the reasoning easier to audit and revise. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

This part of the record rewards slow reading: labels, identifiers and relationships do different jobs. MGnify study accessions commonly begin MGYS. Keeping that accession in notes lets a learner return to the study even when the title is long or similar to another project. The same habit matters beyond school: technicians, analysts, curators and researchers all need to know which layer of a record supports a statement.

Try this: Record both MGnify and archive project identifiers when present, then describe why the pair supports cross-resource tracing. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Compare records only after checking that their organisms, samples, methods, reference versions and study questions are compatible. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

A cheerful surprise in data science is that careful organisation can make a complicated investigation much easier to understand. Sample metadata can describe the organism, habitat, body site, location or experimental condition associated with sequence data. Completeness varies, so environmental context should be read from fields rather than guessed from a title. This is a practical lesson in scientific literacy because structured data can look authoritative even when an essential qualifier is missing.

Try this: Choose two samples and build a context table before looking at any taxonomic chart; stop if the intended comparison is not supported. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. A database result is a route into evidence, not a verdict. Preserve the accession, date and filters so another reader can retrace the route. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Treat this as a small evidence puzzle. The visible fields tell you what can be checked and what still needs another source. A sequencing run accession points to raw sequence output deposited in an archive such as ENA. Multiple runs may belong to one sample, and one study may contain many samples, so run count is not automatically sample count. Reading the field in context helps a learner explain both the useful signal and the limit of the claim.

Try this: Create a run-to-sample mapping for a small subset and calculate both counts separately. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. When the interface changes, the stable identifier and source documentation matter more than the position of a button on the page. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Start with the record itself, because a good science explanation names the object before interpreting it. Assemblies combine sequence reads into longer sequences under a computational process. MGnify can link assembly accessions and analyses, but an assembly is not the untouched raw data and should be described as derived. For secondary science learning, that distinction turns database browsing into evidence practice rather than fact collecting.

Try this: Trace one assembly back toward its source runs, noting every step you can verify and leaving gaps explicit. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Do not fill a blank with a guess. Record the missing field, follow the linked accession or method, and state exactly what remains uncertain. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Here the useful move is to separate what the archive stores from what a researcher may later infer. MGnify analysis records identify pipeline-produced results and commonly use MGYA accessions. A study can have multiple analyses because different inputs or pipeline versions may be processed separately. It also connects classroom habits—defining variables, tracing methods and comparing like with like—to real research infrastructure.

Try this: Compare two analysis records within a study and list only differences visible in their metadata. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Keep observation and interpretation in separate sentences. That simple discipline makes the reasoning easier to audit and revise. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

This part of the record rewards slow reading: labels, identifiers and relationships do different jobs. Computational pipelines perform quality control, classification and other steps according to defined versions and reference resources. The output is reproducible only to the extent that these choices and inputs are recorded. The same habit matters beyond school: technicians, analysts, curators and researchers all need to know which layer of a record supports a statement.

Try this: Write a reproducibility recipe containing accession, analysis type, pipeline version, input and retrieval date. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Compare records only after checking that their organisms, samples, methods, reference versions and study questions are compatible. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

A cheerful surprise in data science is that careful organisation can make a complicated investigation much easier to understand. Taxonomic summaries assign or estimate sequence evidence against reference classifications. A bar or table reports the analysis result, not a direct census of every organism living in the original environment. This is a practical lesson in scientific literacy because structured data can look authoritative even when an essential qualifier is missing.

Try this: Take one taxonomic output and rewrite “these organisms were present” as a more precise statement about classified sequence evidence. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. A database result is a route into evidence, not a verdict. Preserve the accession, date and filters so another reader can retrace the route. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Treat this as a small evidence puzzle. The visible fields tell you what can be checked and what still needs another source. Functional outputs connect sequence features with families, pathways or annotations according to computational rules and databases. They can generate useful hypotheses, but function may be predicted rather than experimentally measured in the sampled environment. Reading the field in context helps a learner explain both the useful signal and the limit of the claim.

Try this: Label three functional statements as measured, computationally assigned or interpretive, then revise any sentence that mixes the categories. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. When the interface changes, the stable identifier and source documentation matter more than the position of a button on the page. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Start with the record itself, because a good science explanation names the object before interpreting it. Microbiome charts often show proportions among classified observations. Relative abundance can change because one group changes or because the denominator changes; it does not automatically reveal absolute cell numbers. For secondary science learning, that distinction turns database browsing into evidence practice rather than fact collecting.

Try this: Create two toy communities with different total counts but the same percentage, then explain why identical proportions need not mean identical abundance. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Do not fill a blank with a guess. Record the missing field, follow the linked accession or method, and state exactly what remains uncertain. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Here the useful move is to separate what the archive stores from what a researcher may later infer. A taxon or function missing from an output may be truly absent, below detection, unclassified, filtered or unsupported by the chosen reference. Careful language protects the distinction. It also connects classroom habits—defining variables, tracing methods and comparing like with like—to real research infrastructure.

Try this: Replace “absent” with a statement tied to the method, such as “not detected in this analysis under the reported pipeline”. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Keep observation and interpretation in separate sentences. That simple discipline makes the reasoning easier to audit and revise. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

This part of the record rewards slow reading: labels, identifiers and relationships do different jobs. Comparisons need compatible sample context, sequencing approach, pipeline and output units. A habitat label alone cannot guarantee that two analyses are comparable. The same habit matters beyond school: technicians, analysts, curators and researchers all need to know which layer of a record supports a statement.

Try this: Build a comparability checklist for two analyses and decide whether comparison is descriptive, exploratory or unsupported. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Compare records only after checking that their organisms, samples, methods, reference versions and study questions are compatible. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

A cheerful surprise in data science is that careful organisation can make a complicated investigation much easier to understand. The MGnify API presents studies, analyses, runs, assemblies, publications and related objects in structured responses. Links between endpoints can support reproducible data exploration and small coding projects. This is a practical lesson in scientific literacy because structured data can look authoritative even when an essential qualifier is missing.

Try this: Retrieve one study or analysis through the API view and identify its type, id, attributes and relationship links. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. A database result is a route into evidence, not a verdict. Preserve the accession, date and filters so another reader can retrace the route. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Treat this as a small evidence puzzle. The visible fields tell you what can be checked and what still needs another source. A linked publication can explain sampling, laboratory methods, analysis choices and biological conclusions. MGnify organises data and analysis outputs, while the paper supplies a narrative that must still be evaluated against the records. Reading the field in context helps a learner explain both the useful signal and the limit of the claim.

Try this: Make a two-column claim audit: publication statement on the left, supporting archive or analysis object on the right. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. When the interface changes, the stable identifier and source documentation matter more than the position of a button on the page. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Start with the record itself, because a good science explanation names the object before interpreting it. Microbiome associations are easily turned into causal or health claims that the study design does not support. Sample size, confounding, compositionality and validation all matter. For secondary science learning, that distinction turns database browsing into evidence practice rather than fact collecting.

Try this: Rewrite a sensational microbiome headline into a claim that names the sample set, analysis and uncertainty. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Do not fill a blank with a guess. Record the missing field, follow the linked accession or method, and state exactly what remains uncertain. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Next chapter

Here the useful move is to separate what the archive stores from what a researcher may later infer. Microbial ecology, public health, food science, environmental monitoring, bioinformatics and data curation all use microbiome evidence. Each role depends on communicating the route from sample to sequence to analysis. It also connects classroom habits—defining variables, tracing methods and comparing like with like—to real research infrastructure.

Try this: Pick one role and design a handover note containing the identifiers and context the next team member would need. Then add one sentence beginning ‘This record can support…’ and another beginning ‘This record cannot by itself establish…’. Keep observation and interpretation in separate sentences. That simple discipline makes the reasoning easier to audit and revise. That two-sentence finish keeps the activity optimistic, precise and honest.

Contents · Previous chapter · Continue at the Science Learning Hub