eduKateSG · Why Science?
Give roots, leaves, tissues and growth stages shared identities—then keep the living plant and experiment in view
Choose whether your question concerns plant structure, development, gene expression or phenotype, then keep species, stage and sampled tissue attached to every comparison.
Science learning becomes powerful when students can name a plant structure, place it in development and connect it to an observation. The Plant Ontology, or PO, is described by the OBO Foundry as a structured vocabulary and database resource linking plant anatomy, morphology, growth and development to plant genomics data. The official page says PO is under active development to encompass terms and annotations from all plants and lists use by Planteome for tissue-expression annotation, as well as examples involving maize and Gramene. The repository distinguishes release files from frequently changing editor files and offers a basic subset limited to is-a and part-of relations. PO makes cross-dataset comparison possible, but an ontology term does not replace species, stage, environment, sample method or measured phenotype. This guide supports Primary Science, PSLE Science, Secondary Science, O-Level Science, O-Level Biology and STEM education as a longform evidence-reading route; it is not agricultural advice, a laboratory protocol, an admission promise or a career guarantee.
Related eduKate reading: Uberon and cross-species anatomy; Phenotype and Trait Ontology and computable qualities; Gene Ontology and traceable gene-function annotations; Environment Ontology and comparable environmental context; Science Learning Hub; Education Hub; STEM Education; Careers by Subject Capability.
Primary and official sources checked on 13 October 2026: OBO Foundry Plant Ontology overview, checked 13 October 2026; Official Plant Ontology source repository; Canonical Plant Ontology OWL release; Planteome plant ontology browser; 2026 Singapore–Cambridge O-Level Biology syllabus; MOE G2/G3 Lower Secondary Science teaching and learning syllabus; 2026 MOE G2 Computing syllabus. Ontologies, definitions and release products can change, so preserve dated documentation, exact source records and file versions.
How plant structure and development become comparable data
Plant words can look familiar while hiding different scales. “Leaf” may mean an organ, a sampled blade, one tissue layer or a developmental state. PO supplies stable classes so the intended level can travel with the data.
The OBO Foundry describes PO as joining plant anatomy, morphology, growth and development with genomics. That connection lets gene-expression or phenotype data be queried by structure and stage rather than by local labels alone.
PO has two major kinds of organisation: anatomical entities and development-stage concepts. A root tissue and a flowering stage answer different questions even when both appear in one experiment.
Part-of relations create a plant partonomy from whole plant through organs, tissues and cells. Query expansion through parts should be deliberate because a sample labelled as an organ can contain several tissues.
Developmental relations link precursors and later structures. The relation supports stage-aware analysis but does not mean every species follows an identical visible sequence.
The official repository separates editor files from release products and warns that the editor version changes frequently. Analyses should use a named release rather than whichever working file happened to be downloaded.
The repository says the release process asserts inferred relations in the release files, while the editor file may not. This means apparent edge counts and query results can depend on product selection.
A basic OBO subset contains only is-a and part-of relations. It is useful for simpler systems but intentionally omits other relationship detail, so limitations should be declared.
Planteome uses PO to describe tissue location for gene expression across green plants. A shared structure term supports search, while species, genotype, developmental stage and assay still shape interpretation.
Cross-species annotation can use common classes without erasing botanical difference. A shared parent may be the right comparison level when specialised structures do not map exactly.
Plant phenotype descriptions often combine a PO entity with a PATO quality. The entity says what is described; the quality says how it differs; measurement and environment complete the evidence.
The optimistic promise is a living comparative map: students and scientists can connect roots, shoots, tissues and stages across datasets while keeping local botanical knowledge visible.
Begin with the sampled material. A whole leaf, leaf blade, epidermis, vascular tissue and single cell can yield very different expression or chemical profiles.
Record the plant’s taxon, cultivar or genotype. Shared anatomy does not make genetic background irrelevant to development, physiology or response.
Growth stage must be observed or operationally defined. Chronological age alone may not align plants grown under different light, temperature or water conditions.
Environment belongs beside anatomy. A root sample from flooded soil and one from dry substrate can share a PO term while expressing very different biology.
Orientation and position matter: proximal, distal, adaxial, abaxial, apical and basal regions can differ within one organ.
A phenotype label should preserve the measurement. “Short root” depends on comparator, unit, imaging method and developmental time.
Images are evidence only when scale, orientation, specimen identity and scoring method remain attached. A cropped photograph can lose the plant context needed to interpret structure.
Gene-expression annotations need assay and sample composition. Bulk organ data may average several cell types, while single-cell data introduce clustering and dissociation biases.
For classroom work, pupils can annotate a bean seedling diagram with structure and stage while clearly separating observed features from inferred function.
Responsible comparison welcomes exceptions. Unexpected structures or stages may reveal error, developmental plasticity or biology that the ontology still needs to represent.
Botanical homology should not be inferred from a shared everyday name. Structures with similar function or appearance can have different developmental and evolutionary histories.
One-to-many mappings are informative. A broad local term may split across several PO classes depending on species, region or stage; forcing one exact mapping hides the ambiguity.
Many-to-one harmonisation can support a large analysis while reducing resolution. Keep the original source term so specialists can revisit the grouped result.
Tissue contamination is a biological and technical issue. Dissection boundaries, adherent cells and transport tissues can make a clean ontology label sound more precise than the sample.
Stage series need repeated observation or well-supported markers. Inferring stage from file order or collection date can create circular conclusions.
Before pooling plant data, build a matrix of PO term, taxon, genotype, stage, tissue region, treatment, environment, assay and mapping confidence. Review mismatches before calculating averages.
Agricultural relevance requires restraint. An ontology-supported association does not guarantee crop performance, safety, yield or suitability under local conditions.
Release migration should flag renamed, merged, split and obsolete classes and show whether figures, counts or model inputs changed.
A reader-ready plant figure should state the PO class and stage in the caption, then give enough plain botanical language to understand the sample without a browser.
PO turns diverse plant observations into connected evidence. Its success is measured not by eliminating variation but by making structure, stage and difference easier to compare honestly.
Experimental replication needs its own botanical map. Several leaves from one plant, several plants from one pot and several pots from one growth chamber represent nested sampling levels, not an interchangeable pile of replicates. Keep the plant, organ, pot, chamber and batch identities separate so the PO annotation remains connected to the statistical design.
Temporal experiments should distinguish chronological time from developmental progress. Two seedlings measured on the same day can occupy different PO stages because germination, temperature and genotype changed their development. Record both clock time and observed stage, then explain which dimension the comparison uses.
| Stage | Question | Evidence | Responsible output |
|---|---|---|---|
| Identify | Which plant structure or stage is meant? | PO identifier, definition, species and stage | A stable anatomy or development class |
| Locate | Where does the structure sit in the plant? | Partonomy, laterality, position and sampled region | A bounded anatomical context |
| Measure | What was observed in that structure or stage? | Trait, expression, image, method and environment | A traceable plant observation |
| Compare | Which cross-species or cross-study conclusion is justified? | Mappings, releases, evidence and validation | A bounded comparison, not automatic equivalence |
Inside this guide
1–12 · Structure, development and identity
- 1. Begin with the plant question
- 2. Use the PO identifier
- 3. Read the definition
- 4. Choose anatomical or stage context
- 5. Inspect the is-a hierarchy
- 6. Inspect part-of relations
- 7. Trace develops-from relations
- 8. Record whole-plant stage
- 9. Record structure development stage
- 10. Record taxon and genotype
- 11. Record sampled region
- 12. Record orientation
13–24 · Genomics, mappings and validation
- 13. Annotate roots precisely
- 14. Annotate shoots precisely
- 15. Annotate leaves precisely
- 16. Annotate reproductive structures
- 17. Connect Gene Ontology annotations
- 18. Connect PATO qualities
- 19. Connect environmental context
- 20. Annotate expression samples
- 21. Compare species at the right level
- 22. Challenge everyday names
- 23. Challenge bulk-sample purity
- 24. Challenge stage inference
25–36 · Learning, agriculture and pathways
- 25. Challenge one-to-many mappings
- 26. Challenge false homology
- 27. Review release products
- 28. Connect to Primary Science
- 29. Connect to Secondary Science
- 30. Connect to mathematics
- 31. Connect to computing
- 32. Connect to pathways
- 33. Did you know?
- 34. Write a reader-ready plant annotation
- 35. Build a validation ladder
- 36. Finish with a reproducible plant-data bundle
Section 1 of 36
1. Begin with the plant question
Anatomy, development, expression, phenotype and ecology need different levels of PO detail.
Working checkpoint — PO identifier: Write the biological claim and unit of observation. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Section 2 of 36
2. Use the PO identifier
A stable identifier connects the class to definitions, synonyms, hierarchy and release history.
Working checkpoint — plant anatomical entity: Store identifier, label and ontology release. Inspect class parents, parts and developmental relations instead of trusting the short label.
For computing, validate taxon constraints, partonomy and obsolete PO identifiers.
Section 3 of 36
3. Read the definition
Familiar botanical labels can hide boundaries, developmental criteria and taxonomic scope.
Working checkpoint — plant structure development stage: Compare the definition with the specimen. Pin the PO release product and record whether inferred relations were included.
For writing, state the sampled structure and stage before discussing mechanism.
Section 4 of 36
4. Choose anatomical or stage context
A structure term and a development-stage term describe different dimensions of the experiment.
Working checkpoint — part-of relation: Record both when the evidence supports them. Preserve original botanical terms and mapping scope for cross-species comparisons.
For students, ask whether the word names a structure, a stage, a quality or a function.
Section 5 of 36
5. Inspect the is-a hierarchy
Broader and narrower plant classes allow analysis at several resolutions.
Working checkpoint — develops-from relation: Use the narrowest term justified by sampling. Separate anatomy identity from measured phenotype, gene expression and proposed function.
For parents, the calm checkpoint is whether the plant sample is described precisely enough to repeat.
Section 6 of 36
6. Inspect part-of relations
Organs contain regions, tissues and cells, but a bulk organ sample need not represent every part equally.
Working checkpoint — Planteome annotation: Declare relation-aware query expansion. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Section 7 of 36
7. Trace develops-from relations
Plant structures emerge through developmental processes that can vary with species and condition.
Working checkpoint — PO identifier: Link precursor claims to stage evidence. Inspect class parents, parts and developmental relations instead of trusting the short label.
For computing, validate taxon constraints, partonomy and obsolete PO identifiers.
Section 8 of 36
8. Record whole-plant stage
Germination, vegetative growth, flowering and senescence shape how an anatomical observation should be interpreted.
Working checkpoint — plant anatomical entity: Use observed stage criteria, not date alone. Pin the PO release product and record whether inferred relations were included.
For writing, state the sampled structure and stage before discussing mechanism.
Section 9 of 36
9. Record structure development stage
Individual organs can have their own developmental series within the whole plant’s life cycle.
Working checkpoint — plant structure development stage: Keep organ and whole-plant stage separate. Preserve original botanical terms and mapping scope for cross-species comparisons.
For students, ask whether the word names a structure, a stage, a quality or a function.
Section 10 of 36
10. Record taxon and genotype
Species, cultivar and genotype influence morphology, timing and response even under shared PO classes.
Working checkpoint — part-of relation: Attach stable taxonomic and genotype records. Separate anatomy identity from measured phenotype, gene expression and proposed function.
For parents, the calm checkpoint is whether the plant sample is described precisely enough to repeat.
Section 11 of 36
11. Record sampled region
Blade, base, tip, surface and internal tissue can differ within one named organ.
Working checkpoint — develops-from relation: Add positional qualifiers and dissection notes. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Section 12 of 36
12. Record orientation
Adaxial, abaxial, proximal, distal, apical and basal positions support distinct observations.
Working checkpoint — Planteome annotation: Preserve orientation in text and structured fields. Inspect class parents, parts and developmental relations instead of trusting the short label.
For computing, validate taxon constraints, partonomy and obsolete PO identifiers.
Section 13 of 36
13. Annotate roots precisely
Root systems contain zones, tissues and developmental states that should not collapse into one generic root label.
Working checkpoint — PO identifier: Match term depth to sampling resolution. Pin the PO release product and record whether inferred relations were included.
For writing, state the sampled structure and stage before discussing mechanism.
Section 14 of 36
14. Annotate shoots precisely
Stem, node, internode, bud and meristem terms support different developmental and expression questions.
Working checkpoint — plant anatomical entity: Keep spatial and stage context visible. Preserve original botanical terms and mapping scope for cross-species comparisons.
For students, ask whether the word names a structure, a stage, a quality or a function.
Section 15 of 36
15. Annotate leaves precisely
Leaf organ, blade, petiole, margin, epidermis and vascular tissue are not interchangeable samples.
Working checkpoint — plant structure development stage: Name the actual material measured. Separate anatomy identity from measured phenotype, gene expression and proposed function.
For parents, the calm checkpoint is whether the plant sample is described precisely enough to repeat.
Section 16 of 36
16. Annotate reproductive structures
Flowers, inflorescences, pollen, ovules, fruits and seeds require clear structural and stage identity.
Working checkpoint — part-of relation: Record reproductive stage and parent structure. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Section 17 of 36
17. Connect Gene Ontology annotations
PO can supply anatomical context for where gene products or processes are observed.
Working checkpoint — develops-from relation: Separate location evidence from functional interpretation. Inspect class parents, parts and developmental relations instead of trusting the short label.
For computing, validate taxon constraints, partonomy and obsolete PO identifiers.
Section 18 of 36
18. Connect PATO qualities
Entity–quality descriptions can pair a PO structure with size, shape, colour or other qualities.
Working checkpoint — Planteome annotation: Retain comparator, value, unit and method. Pin the PO release product and record whether inferred relations were included.
For writing, state the sampled structure and stage before discussing mechanism.
Section 19 of 36
19. Connect environmental context
Light, water, temperature, substrate and treatment influence plant state without changing the anatomy identifier.
Working checkpoint — PO identifier: Link ENVO context and measured conditions. Preserve original botanical terms and mapping scope for cross-species comparisons.
For students, ask whether the word names a structure, a stage, a quality or a function.
Section 20 of 36
20. Annotate expression samples
Planteome and other resources use PO to describe tissue context for plant genomic data.
Working checkpoint — plant anatomical entity: Keep assay, taxon, genotype and stage. Separate anatomy identity from measured phenotype, gene expression and proposed function.
For parents, the calm checkpoint is whether the plant sample is described precisely enough to repeat.
Section 21 of 36
21. Compare species at the right level
A shared parent can support comparison when specialised structures lack exact equivalence.
Working checkpoint — plant structure development stage: State the level and mapping relation. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Section 22 of 36
22. Challenge everyday names
Common words can refer to different botanical structures across species or local traditions.
Working checkpoint — part-of relation: Prefer definitions and source terminology over string matching. Inspect class parents, parts and developmental relations instead of trusting the short label.
For computing, validate taxon constraints, partonomy and obsolete PO identifiers.
Section 23 of 36
23. Challenge bulk-sample purity
A dissected organ can contain several tissues and cell types in unequal proportions.
Working checkpoint — develops-from relation: Record method and composition limits. Pin the PO release product and record whether inferred relations were included.
For writing, state the sampled structure and stage before discussing mechanism.
Section 24 of 36
24. Challenge stage inference
Collection time, size or colour may be imperfect proxies for developmental stage.
Working checkpoint — Planteome annotation: Use explicit operational criteria and uncertainty. Preserve original botanical terms and mapping scope for cross-species comparisons.
For students, ask whether the word names a structure, a stage, a quality or a function.
Section 25 of 36
25. Challenge one-to-many mappings
One legacy label can map to several PO terms depending on region or species.
Working checkpoint — PO identifier: Preserve alternatives and decision rules. Separate anatomy identity from measured phenotype, gene expression and proposed function.
For parents, the calm checkpoint is whether the plant sample is described precisely enough to repeat.
Section 26 of 36
26. Challenge false homology
Similarity of shape, position or function does not alone establish common evolutionary origin.
Working checkpoint — plant anatomical entity: Use separate homology evidence for evolutionary claims. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Section 27 of 36
27. Review release products
Editor, release and basic files contain different stability and relation content.
Working checkpoint — plant structure development stage: Name the exact product and checksum. Inspect class parents, parts and developmental relations instead of trusting the short label.
For computing, validate taxon constraints, partonomy and obsolete PO identifiers.
Section 28 of 36
28. Connect to Primary Science
Seeds, roots, stems, leaves and life cycles make structure and stage visible through direct observation.
Working checkpoint — part-of relation: Use safe plants and dated drawings. Pin the PO release product and record whether inferred relations were included.
For writing, state the sampled structure and stage before discussing mechanism.
Section 29 of 36
29. Connect to Secondary Science
Transport, photosynthesis, reproduction and response depend on structures studied at several levels.
Working checkpoint — develops-from relation: Link every functional claim to observed anatomy. Preserve original botanical terms and mapping scope for cross-species comparisons.
For students, ask whether the word names a structure, a stage, a quality or a function.
Section 30 of 36
30. Connect to mathematics
Growth curves, allometry, sampling and variability show why stage and replicate structure matter.
Working checkpoint — Planteome annotation: Plot invented data by stage rather than age alone. Separate anatomy identity from measured phenotype, gene expression and proposed function.
For parents, the calm checkpoint is whether the plant sample is described precisely enough to repeat.
Section 31 of 36
31. Connect to computing
Ontology graphs, identifiers, releases and mappings make plant observations queryable across databases.
Working checkpoint — PO identifier: Test obsolete IDs and impossible part relations. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Section 32 of 36
32. Connect to pathways
PO connects botany, ecology, agriculture, genomics, imaging and scientific data stewardship.
Working checkpoint — plant anatomical entity: Check current official study routes separately. Inspect class parents, parts and developmental relations instead of trusting the short label.
For computing, validate taxon constraints, partonomy and obsolete PO identifiers.
Section 33 of 36
33. Did you know?
The official repository provides a basic PO subset limited to is-a and part-of relations, while full release files include a richer relation model.
Working checkpoint — plant structure development stage: Ask which product a database used before comparing graph results. Pin the PO release product and record whether inferred relations were included.
For writing, state the sampled structure and stage before discussing mechanism.
Section 34 of 36
34. Write a reader-ready plant annotation
Name PO class, source term, taxon, genotype, stage, region, environment, method, release and uncertainty.
Working checkpoint — part-of relation: Put the botanical evidence before the interpretation. Preserve original botanical terms and mapping scope for cross-species comparisons.
For students, ask whether the word names a structure, a stage, a quality or a function.
Section 35 of 36
35. Build a validation ladder
Check identifier, definition, anatomy, stage, taxon, sampling, phenotype, expert review and replication in order.
Working checkpoint — develops-from relation: Stop the conclusion at the highest completed rung. Separate anatomy identity from measured phenotype, gene expression and proposed function.
For parents, the calm checkpoint is whether the plant sample is described precisely enough to repeat.
Section 36 of 36
36. Finish with a reproducible plant-data bundle
Deliver samples, PO mappings, releases, images, measurements, environments, code, tests and limitations.
Working checkpoint — Planteome annotation: Make every plant comparison traceable to structure and stage. Keep PO term, taxon, genotype, stage, sampled region, method and environment together.
For mathematics, keep growth curves, units, stages and biological replicates visible.
Contents · Previous section · Continue to the Science Learning Hub