eduKateSG · Why Science?
Extend familiar FASTA sequences with computable database provenance, variants, modifications and processing events—without losing the original protein record
Choose the closest route, then return to the full index whenever you need the wider evidence chain.
Science learning becomes powerful when students can tell a measured pattern from a model, a statistical decision from a mechanism, and a useful lead from a finished conclusion. PEFF, the PSI Extended FASTA Format, is a ratified HUPO Proteomics Standards Initiative standard for protein and nucleotide sequence databases used by sequence search engines and related tools. The official page says it extends FASTA-style entries with a file header describing source database name and version, consistent identifiers and descriptions, taxonomy and structured annotations such as post-translational modifications, sequence variants and processing events. PEFF can make a search database richer and more portable, but an annotation in a database is not proof that a particular sample contains that proteoform or that a mass-spectrum match is correct. This guide supports Primary Science, PSLE Science, Secondary Science, O-Level Science and STEM education as a longform evidence-reading route; it is not medical advice, a laboratory protocol, an admission promise or a career guarantee.
Related eduKate reading: mzIdentML and proteomics identification results; ProForma and proteoform sequence notation; mass spectrometry, ionisation and mass-to-charge evidence; Science Learning Hub; Education Hub; STEM Education; Careers by Subject Capability.
Primary and official sources checked on 12 October 2026: HUPO-PSI PEFF official page and ratified-standard materials, checked 12 October 2026; HUPO-PSI PEFF official repository and specification documents, checked 12 October 2026; HUPO-PSI PEFF peer-reviewed standard article, PMID 31081335, checked 12 October 2026; 2026 Singapore–Cambridge O-Level Biology syllabus; MOE G2/G3 Lower Secondary Science teaching and learning syllabus; 2026 MOE G2 Computing syllabus. Specifications, vocabularies, tools and recommendations can change, so preserve dated documentation, exact source data and software versions.
A deeper reading route
Classic FASTA is successful because it is simple: a header line names an entry and following lines carry sequence. Its descriptive header, however, varies widely among providers and tools. PEFF keeps the familiar plain-text sequence model while defining metadata and entry annotations so software does not have to guess what each provider-specific phrase means.
A PEFF file begins with database-level metadata. Name, release, source and prefix information identify the sequence collection and define how entry identifiers should be read. Search results are only reproducible when the exact database bytes or a verifiable release remain available; the same accession can acquire new sequence or annotation in a later release.
Entry identifiers need namespaces and stable parsing. A reader should distinguish the database prefix, local identifier, accession and display description rather than treating the whole header as an unstructured name. That separation improves cross-tool exchange and prevents accidental joins between similarly labelled records.
Structured key-value annotations can represent sequence variants, modified residues, processed chains, initiator methionine removal and other events. Positions are tied to the underlying sequence. Inserting, deleting or changing sequence without remapping annotations can move every downstream coordinate while leaving plausible-looking text.
Known variants can expand proteogenomic searches, but a catalogue variant is not the same as a sample genotype. Search spaces grow as alternatives are added, affecting sensitivity, false-discovery control and computational cost. Reports should state which variants were enabled, their sources and how variant peptides were validated.
Modification annotations can restrict or describe known sites. A database entry may say that a residue can be modified, yet the searched spectrum must still support the modified peptide and its localisation. Database knowledge guides hypotheses; it does not replace experimental evidence.
Processing events can describe signal peptides, propeptides, mature chains or cleavage products. These annotations improve biological context, but coordinate systems must remain explicit. A peptide located in a mature chain may use positions relative to the full precursor sequence in the database.
PEFF can represent multiple source databases in a governed file. Prefixes and database descriptions keep records distinct. Merging sequence collections without preserving their provenance can create identifier collisions and duplicate sequences that distort inference or search statistics.
Search engines and converters support different PEFF drafts or features. The official implementation table records support dates and levels rather than promising universal compatibility. A laboratory should test the exact annotations it uses with small fixtures and record any fields that a tool ignores.
Validation should inspect the file header, database declarations, entry syntax, allowed keys, escaping, positions, residue compatibility and unique identifiers. A valid PEFF file can still contain outdated or scientifically incorrect annotations, so content provenance and release policy need a second review layer.
Decoy generation deserves care. Search workflows often combine target and decoy sequences for error estimation. Variants and modifications can change search-space structure. Preserve the generator, strategy, seed where relevant and final database hash so error calculations can be audited.
PEFF and ProForma solve related but different problems. PEFF describes sequence-database entries and their possible annotations; ProForma writes a specific peptidoform or proteoform sequence string. A measured identification may use a PEFF database and report the supported molecular form in ProForma.
For students, PEFF is like turning a simple list of words into a dictionary whose entries have reliable labels, origins, spelling variants and notes. The extra structure lets computers use the notes without pretending every possible note applies in every sentence.
A classroom model can use fictional protein letter strings and annotation cards. Students add a variant, modified site and processed chain, then test what happens after one residue is inserted. They quickly see why positions, versions and provenance are scientific data.
A reproducible search bundle includes the exact PEFF bytes and hash, specification, source-database releases, converters, enabled annotation types, search-engine version, parameters, decoy strategy, identification results and validation logs. This preserves the bridge from database hypothesis to spectral claim.
The optimistic value of PEFF is portable richness. Sequence providers can share more than letters without trapping meaning in local prose. When tools and reports preserve version, coordinates and evidence limits, richer databases support more precise and inspectable proteomics.
A safe implementation tests each PEFF feature used by the search engine. Construct tiny records containing one variant, modified residue, processed chain and escaped value; search or parse them and compare the resulting coordinates and annotations. A tool claiming PEFF support may legitimately implement only part of a draft or release, so feature-level evidence is more useful than a logo.
Database growth affects statistics. Adding known variants and modifications increases candidate sequences and peptide forms, which can improve discovery while also increasing competing matches. Error estimation, score calibration and validation thresholds should be evaluated under the actual expanded database, not borrowed from a smaller FASTA search.
When source databases update, a reproducible workflow should build a new PEFF artifact rather than silently replacing an old one. Store source release identifiers, converter version, enabled fields, build log and hash. Differences between builds can then be attributed to sequence or annotation changes instead of mysterious search drift.
A publication should state which PEFF annotations were allowed to influence the search and how resulting peptides were confirmed. That sentence protects a crucial boundary: structured database knowledge suggests hypotheses, while spectra and statistical controls determine which hypotheses receive support.
Duplicate sequences need deliberate handling. Two accessions can legitimately share one protein sequence while representing different genes, isoforms or source databases. Collapsing them may simplify search space but can erase provenance; retaining all of them can complicate protein inference. Record the policy and keep a mapping from search entries to source records.
Coordinate-aware annotations should be tested after every sequence transformation. Removing an initiator residue, selecting a mature chain or applying a variant can create a new coordinate frame. The original full-sequence position and transformed search-sequence position should both remain available when peptide results are reported.
Text encoding and line wrapping are practical preservation issues. A PEFF-aware writer should follow the specification for characters, delimiters and sequence layout, while transfer tools should avoid changing newlines or wrapping headers. Validate the final delivered bytes, not only an in-memory representation before export.
Rich annotations are most valuable when readers can see what was ignored. A search report should list supported keys, unsupported keys and any converter warnings. This turns partial implementation from a hidden risk into an explicit, testable boundary.
The final database note should also record whether nucleotide entries, protein entries or both were present and which alphabet rules applied. That small declaration prevents a downstream tool from treating an unexpected sequence type as corruption or silently skipping it, and it completes the chain from collection scope to search result.
| Stage | Question | Evidence | Responsible output |
|---|---|---|---|
| Identify | Which database and release supplied the entry? | File header, prefix, source and version | A reproducible collection |
| Sequence | What base string is searched? | Entry identifier and amino-acid sequence | A reference protein |
| Annotate | Which alternatives or events are known? | Variants, modifications and processing keys | Structured hypotheses |
| Test | What does the experiment support? | Search settings, spectra, scores and validation | A bounded identification |
Inside this guide
1–12 · Foundations and design
- 1. Start with the database question
- 2. Recognise a ratified standard
- 3. Preserve the exact file
- 4. Read the PEFF header
- 5. Read database prefixes
- 6. Read entry identifiers
- 7. Read descriptions
- 8. Read taxonomy
- 9. Read the base sequence
- 10. Read simple variants
- 11. Read complex variants
- 12. Read modified residues
13–24 · Models and estimation
- 13. Read processing events
- 14. Read uncertain annotations
- 15. Read cross-references
- 16. Read multiple databases
- 17. Check escaped text
- 18. Check coordinate ranges
- 19. Validate file syntax
- 20. Validate annotation semantics
- 21. Test tool support
- 22. Test round trips
- 23. Record search settings
- 24. Record decoy strategy
25–36 · Diagnostics and interpretation
- 25. Protect sensitive variants
- 26. Separate database knowledge from observation
- 27. Separate sequence from function
- 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 database note
- 35. Build a validation ladder
- 36. Finish with a reproducible bundle
Section 1 of 36
1. Start with the database question
Name the search purpose, organism, source databases and required annotations.
Working checkpoint — PEFF: Avoid an uncontrolled sequence mixture. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Section 2 of 36
2. Recognise a ratified standard
The official PEFF page says the specification completed PSI review and was released.
Working checkpoint — PSI Extended FASTA Format: Use the released documents. Read every annotation beside its sequence position, source, database version and supported tool feature.
For future study, connect proteins to databases and algorithms while checking official pathways separately.
Section 3 of 36
3. Preserve the exact file
Database release names are helpful, but byte hashes make the search input verifiable.
Working checkpoint — protein sequence database: Archive file and checksum. Preserve alternatives and obsolete history; a richer database is a hypothesis space, not a sample observation.
For scientific writing, cite PEFF release, database sources, search settings, results and validator.
Section 4 of 36
4. Read the PEFF header
Database-level metadata defines sources, versions, prefixes and format context.
Working checkpoint — sequence variant: Resolve every declaration. Separate searchable knowledge from spectral identification, protein function and clinical meaning.
For students, ask which sequence, position and source support each annotation.
Section 5 of 36
5. Read database prefixes
Prefixes separate identifiers from different source collections.
Working checkpoint — post-translational modification: Prevent namespace collisions. Treat identifiers, escaping and coordinates as scientific inputs: one sequence edit can move every annotation.
For parents, ask whether a rich database keeps its version and evidence limits visible.
Section 6 of 36
6. Read entry identifiers
Stable identifiers support search-result links and updates.
Working checkpoint — proteogenomics: Do not parse descriptions as IDs. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Section 7 of 36
7. Read descriptions
Human-readable names help readers but may change or be non-unique.
Working checkpoint — PEFF: Join on governed identifiers. Read every annotation beside its sequence position, source, database version and supported tool feature.
For future study, connect proteins to databases and algorithms while checking official pathways separately.
Section 8 of 36
8. Read taxonomy
Organism context is part of sequence-database meaning.
Working checkpoint — PSI Extended FASTA Format: Verify taxonomy source and scope. Preserve alternatives and obsolete history; a richer database is a hypothesis space, not a sample observation.
For scientific writing, cite PEFF release, database sources, search settings, results and validator.
Section 9 of 36
9. Read the base sequence
All annotations depend on the exact stored residue order.
Working checkpoint — protein sequence database: Validate alphabet and length. Separate searchable knowledge from spectral identification, protein function and clinical meaning.
For students, ask which sequence, position and source support each annotation.
Section 10 of 36
10. Read simple variants
A variant needs position, original context and alternative residue.
Working checkpoint — sequence variant: Check residue compatibility. Treat identifiers, escaping and coordinates as scientific inputs: one sequence edit can move every annotation.
For parents, ask whether a rich database keeps its version and evidence limits visible.
Section 11 of 36
11. Read complex variants
Insertions, deletions and substitutions alter downstream coordinates.
Working checkpoint — post-translational modification: Follow the specification carefully. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Section 12 of 36
12. Read modified residues
Known modifications can annotate candidate sites.
Working checkpoint — proteogenomics: Separate database possibility from measured evidence. Read every annotation beside its sequence position, source, database version and supported tool feature.
For future study, connect proteins to databases and algorithms while checking official pathways separately.
Section 13 of 36
13. Read processing events
Signal peptides, chains and cleavage events describe derived forms.
Working checkpoint — PEFF: Keep full-sequence coordinates. Preserve alternatives and obsolete history; a richer database is a hypothesis space, not a sample observation.
For scientific writing, cite PEFF release, database sources, search settings, results and validator.
Section 14 of 36
14. Read uncertain annotations
Evidence strength and ambiguity should remain visible when represented.
Working checkpoint — PSI Extended FASTA Format: Do not upgrade uncertainty. Separate searchable knowledge from spectral identification, protein function and clinical meaning.
For students, ask which sequence, position and source support each annotation.
Section 15 of 36
15. Read cross-references
External accessions need source databases and stable mappings.
Working checkpoint — protein sequence database: Pin releases. Treat identifiers, escaping and coordinates as scientific inputs: one sequence edit can move every annotation.
For parents, ask whether a rich database keeps its version and evidence limits visible.
Section 16 of 36
16. Read multiple databases
One PEFF file can describe governed source collections.
Working checkpoint — sequence variant: Preserve every provenance block. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Section 17 of 36
17. Check escaped text
Delimiters and backslashes have syntactic roles in structured headers.
Working checkpoint — post-translational modification: Use a PEFF-aware parser. Read every annotation beside its sequence position, source, database version and supported tool feature.
For future study, connect proteins to databases and algorithms while checking official pathways separately.
Section 18 of 36
18. Check coordinate ranges
Every annotated position must exist in the corresponding sequence.
Working checkpoint — proteogenomics: Reject out-of-range values. Preserve alternatives and obsolete history; a richer database is a hypothesis space, not a sample observation.
For scientific writing, cite PEFF release, database sources, search settings, results and validator.
Section 19 of 36
19. Validate file syntax
The specification and validator check header and entry grammar.
Working checkpoint — PEFF: Keep validator version and report. Separate searchable knowledge from spectral identification, protein function and clinical meaning.
For students, ask which sequence, position and source support each annotation.
Section 20 of 36
20. Validate annotation semantics
Keys, values, residues and positions must make scientific sense.
Working checkpoint — PSI Extended FASTA Format: Review beyond grammar. Treat identifiers, escaping and coordinates as scientific inputs: one sequence edit can move every annotation.
For parents, ask whether a rich database keeps its version and evidence limits visible.
Section 21 of 36
21. Test tool support
Search engines may implement only selected PEFF features or drafts.
Working checkpoint — protein sequence database: Use a feature fixture. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Section 22 of 36
22. Test round trips
Converters can drop unknown annotations or reorder fields.
Working checkpoint — sequence variant: Compare structured meaning. Read every annotation beside its sequence position, source, database version and supported tool feature.
For future study, connect proteins to databases and algorithms while checking official pathways separately.
Section 23 of 36
23. Record search settings
Variant and modification use expands the hypothesis space.
Working checkpoint — post-translational modification: Preserve enabled fields and tolerances. Preserve alternatives and obsolete history; a richer database is a hypothesis space, not a sample observation.
For scientific writing, cite PEFF release, database sources, search settings, results and validator.
Section 24 of 36
24. Record decoy strategy
Error estimation depends on the exact target-decoy database.
Working checkpoint — proteogenomics: Archive generator and hash. Separate searchable knowledge from spectral identification, protein function and clinical meaning.
For students, ask which sequence, position and source support each annotation.
Section 25 of 36
25. Protect sensitive variants
Custom sample-derived sequences may expose governed genomic information.
Working checkpoint — PEFF: Apply consent and access controls. Treat identifiers, escaping and coordinates as scientific inputs: one sequence edit can move every annotation.
For parents, ask whether a rich database keeps its version and evidence limits visible.
Section 26 of 36
26. Separate database knowledge from observation
A PEFF annotation proposes searchable context.
Working checkpoint — PSI Extended FASTA Format: Require spectrum evidence. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Section 27 of 36
27. Separate sequence from function
A named or modified sequence does not prove mechanism.
Working checkpoint — protein sequence database: Use functional experiments separately. Read every annotation beside its sequence position, source, database version and supported tool feature.
For future study, connect proteins to databases and algorithms while checking official pathways separately.
Section 28 of 36
28. Connect to Primary Science
A labelled sequence card shows why versions and positions matter.
Working checkpoint — sequence variant: Use fictional letters. Preserve alternatives and obsolete history; a richer database is a hypothesis space, not a sample observation.
For scientific writing, cite PEFF release, database sources, search settings, results and validator.
Section 29 of 36
29. Connect to Secondary Science
Variants and processing support lessons about proteins, evidence and uncertainty.
Working checkpoint — post-translational modification: Trace one annotation. Separate searchable knowledge from spectral identification, protein function and clinical meaning.
For students, ask which sequence, position and source support each annotation.
Section 30 of 36
30. Connect to mathematics
Coordinates, insertions and search-space growth build discrete reasoning.
Working checkpoint — proteogenomics: Remap one position. Treat identifiers, escaping and coordinates as scientific inputs: one sequence edit can move every annotation.
For parents, ask whether a rich database keeps its version and evidence limits visible.
Section 31 of 36
31. Connect to computing
Namespaced keys, escaping and validators show structured text design.
Working checkpoint — PEFF: Test one malformed header. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Section 32 of 36
32. Connect to pathways
Sequence databases join biology, chemistry, statistics and software and can illuminate STEM education and career pathways.
Working checkpoint — PSI Extended FASTA Format: Verify current routes separately. Read every annotation beside its sequence position, source, database version and supported tool feature.
For future study, connect proteins to databases and algorithms while checking official pathways separately.
Section 33 of 36
33. Did you know?
PEFF keeps a FASTA-like text sequence while adding database-level metadata and computable entry annotations.
Working checkpoint — protein sequence database: Preserve both layers. Preserve alternatives and obsolete history; a richer database is a hypothesis space, not a sample observation.
For scientific writing, cite PEFF release, database sources, search settings, results and validator.
Section 34 of 36
34. Write a reader-ready database note
State PEFF release, sources, versions, enabled annotations, tool support and limits.
Working checkpoint — sequence variant: Place it beside the database. Separate searchable knowledge from spectral identification, protein function and clinical meaning.
For students, ask which sequence, position and source support each annotation.
Section 35 of 36
35. Build a validation ladder
Check header, namespaces, sequences, keys, coordinates, tools and search receipts.
Working checkpoint — post-translational modification: Stop at the first incompatible annotation. Treat identifiers, escaping and coordinates as scientific inputs: one sequence edit can move every annotation.
For parents, ask whether a rich database keeps its version and evidence limits visible.
Section 36 of 36
36. Finish with a reproducible bundle
Deliver PEFF, source releases, converters, validators, search settings, results and hashes.
Working checkpoint — proteogenomics: Make every peptide traceable. Keep database header, release, prefixes, sequences, annotations, search settings and validation together.
For a classroom model, annotate fictional sequences and remap after one insertion.
Contents · Previous section · Continue to the Science Learning Hub