eduKateSG · Why Science?
Read neighbouring variants together on each haplotype—because two changes can rewrite one codon or rescue one frame
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. BCFtools/csq is a fast consequence caller designed to use known phase so nearby variants on the same haplotype can be interpreted together. The official guide highlights cases where adjacent changes alter the same codon or where one frameshifting indel is followed by a frame-restoring indel. It accepts VCF or BCF variants, a reference FASTA and Ensembl GFF3 features, then records consequences in the output. Haplotype awareness corrects an important modelling weakness, but its conclusions still depend on trustworthy genotype, phase, transcript and reference information. 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: Ensembl VEP and transcript-aware variant consequences; SnpEff and genome-scale variant impact annotation; RegTools and variant-aware splice-junction analysis; Science Learning Hub; Education Hub; STEM Education; Careers by Subject Capability.
Primary and official sources checked on 11 October 2026: Official BCFtools consequence-calling guide; Official BCFtools manual; BCFtools/csq primary study full text; BCFtools/csq primary study record in PubMed; 2026 Singapore–Cambridge O-Level Biology syllabus; MOE G2/G3 Lower Secondary Science teaching and learning syllabus; 2026 MOE G2 Computing syllabus. Package interfaces and recommendations can change, so reproduce an analysis only with the documentation for the installed release.
| Stage | Question | Evidence | Responsible output |
|---|---|---|---|
| Phase | Which alleles occur together on each chromosome copy? | Phased genotypes and sample identities | A declared haplotype context |
| Project | How do those alleles alter a transcript sequence together? | Reference FASTA and Ensembl GFF3 | A reconstructed coding or splice consequence |
| Encode | Where are compound consequences stored? | BCSQ annotations and linked positions | A traceable haplotype-aware record |
| Validate | Does independent evidence support the predicted effect? | RNA, protein and functional measurements | A bounded biological conclusion |
Inside this guide
1–12 · Foundations and design
- 1. Start with haplotypes
- 2. See the isolated-record problem
- 3. Define a compound variant
- 4. Prepare a reference FASTA
- 5. Prepare Ensembl GFF3 features
- 6. Prepare VCF or BCF input
- 7. Read genotype separators
- 8. Understand known phase
- 9. Understand local consequence mode
- 10. Read the BCSQ field
- 11. Follow linked positions
- 12. Combine adjacent substitutions
13–24 · Models and estimation
- 13. Combine variants across an intron
- 14. Recognise frame restoration
- 15. Recognise same-codon rescue
- 16. Respect strand orientation
- 17. Separate phase from proximity
- 18. Handle unphased genotypes cautiously
- 19. Inspect phasing switches
- 20. Check sample identity
- 21. Test multiallelic sites
- 22. Test overlapping indels
- 23. Inspect splice consequences
- 24. Compare with VEP or SnpEff
25–36 · Diagnostics and interpretation
- 25. Read performance claims in context
- 26. Audit one compound consequence
- 27. Use bounded clinical language
- 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 the result responsibly
- 35. Build a validation ladder
- 36. Finish with a reproducible bundle
Section 1 of 36
1. Start with haplotypes
A diploid genome carries two chromosome copies, and nearby alleles may occur together in cis or on opposite copies in trans. Their combined sequence determines the consequence.
Working checkpoint — haplotype-aware consequence: State whether phase is measured, inferred or unknown. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 2 of 36
2. See the isolated-record problem
Many annotators process each VCF record independently. That can misread adjacent substitutions that jointly form one codon or indels whose combined length restores a frame.
Working checkpoint — phase information: Flag clustered variants before interpreting single-record labels. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 3 of 36
3. Define a compound variant
A compound consequence arises when two or more nearby alleles jointly change the transcript interpretation. The unit of inference is the haplotype sequence, not an isolated row.
Working checkpoint — compound variant: Keep all contributing positions linked in the report. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 4 of 36
4. Prepare a reference FASTA
BCFtools/csq reconstructs sequence consequences against the supplied reference. The FASTA assembly and contig names must match both variants and features.
Working checkpoint — GFF3 transcript: Checksum the FASTA and test reference alleles. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 5 of 36
5. Prepare Ensembl GFF3 features
The official guide states that the caller uses Ensembl GFF3 genomic features. Transcript, exon and coding annotations define how genomic alleles project into sequence.
Working checkpoint — BCSQ field: Record the Ensembl release and GFF3 checksum. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 6 of 36
6. Prepare VCF or BCF input
Variant records need valid alleles, genotypes and samples. Normalisation and multiallelic representation affect which events can be combined.
Working checkpoint — functional validation: Archive original and prepared files with transformation commands. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 7 of 36
7. Read genotype separators
In VCF, a vertical bar indicates phased genotype alleles while a slash indicates unphased alleles. The symbol carries biological modelling meaning.
Working checkpoint — haplotype-aware consequence: Reject accidental phase loss during file conversion. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 8 of 36
8. Understand known phase
Known phase tells the caller which alternatives belong on the same haplotype. Incorrect phase can create a compound protein sequence that no chromosome actually carries.
Working checkpoint — phase information: Record the phasing method and confidence. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 9 of 36
9. Understand local consequence mode
The documentation exposes a local mode that disables haplotype-aware combination. It can support comparison with conventional record-wise annotation.
Working checkpoint — compound variant: Run both modes on a test set to identify compound differences. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 10 of 36
10. Read the BCSQ field
BCFtools/csq writes structured consequence information to a BCSQ annotation field and can link compound effects across records. The header defines the schema.
Working checkpoint — GFF3 transcript: Preserve the full header and parse fields programmatically. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 11 of 36
11. Follow linked positions
For compound events spanning several records, the detailed annotation may be attached to one record while others point to it. Reading rows independently can lose the connection.
Working checkpoint — BCSQ field: Expand linked records during review and export. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 12 of 36
12. Combine adjacent substitutions
Two neighbouring SNVs on the same haplotype can form a multiple-nucleotide change and encode a different amino acid than two isolated predictions suggest.
Working checkpoint — functional validation: Reconstruct the codon before assigning protein consequence. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 13 of 36
13. Combine variants across an intron
Variants separated in genomic space can become adjacent in the spliced transcript. Haplotype-aware transcript projection must respect exon order and strand.
Working checkpoint — haplotype-aware consequence: Audit transcript coordinates and spliced sequence. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 14 of 36
14. Recognise frame restoration
One indel may appear frameshifting alone while a later indel restores the reading frame. The intervening peptide can still change even when the downstream frame returns.
Working checkpoint — phase information: Report both the altered segment and the restored frame. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 15 of 36
15. Recognise same-codon rescue
Two substitutions can transform a codon jointly so that an apparent stop or missense from one allele is changed by the second.
Working checkpoint — compound variant: Compare reference, isolated and combined codons. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 16 of 36
16. Respect strand orientation
Genes on the reverse strand require reverse-complement logic when projecting genomic alleles into transcripts. A strand error changes codons and splice boundaries.
Working checkpoint — GFF3 transcript: Hand-check one reverse-strand compound example. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 17 of 36
17. Separate phase from proximity
Nearby alleles are not necessarily on the same chromosome copy. Distance alone cannot justify combining them.
Working checkpoint — BCSQ field: Use explicit genotype phase or report alternative haplotype scenarios. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 18 of 36
18. Handle unphased genotypes cautiously
When phase is unknown, several haplotype arrangements may be compatible with the genotype. A single deterministic consequence can overstate certainty.
Working checkpoint — functional validation: Preserve ambiguity and seek read-backed or pedigree phase. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 19 of 36
19. Inspect phasing switches
Statistical or read-backed phasing can contain switch errors that place downstream alleles on the wrong haplotype. Long compound regions are especially sensitive.
Working checkpoint — haplotype-aware consequence: Use confidence blocks and independent evidence for critical cases. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 20 of 36
20. Check sample identity
Phase and genotype belong to a particular sample. Merging records across sample columns or specimen swaps can manufacture impossible haplotypes.
Working checkpoint — phase information: Verify sample maps and genotype completeness. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 21 of 36
21. Test multiallelic sites
Several alternatives at one position interact with phased genotype indices. Splitting records can preserve or damage phase depending on the transformation.
Working checkpoint — compound variant: Regression-test multiallelic examples after normalisation. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 22 of 36
22. Test overlapping indels
Overlapping or nearby indels can have ambiguous representations and complex sequence outcomes. Standard left alignment may not fully express biological equivalence.
Working checkpoint — GFF3 transcript: Reconstruct the full local haplotype sequence for review. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 23 of 36
23. Inspect splice consequences
Transcript-aware consequence calling can label splice-region effects, but a positional label does not demonstrate altered RNA in the sample.
Working checkpoint — BCSQ field: Add junction or targeted transcript evidence. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 24 of 36
24. Compare with VEP or SnpEff
Traditional annotations remain useful for broad compatibility, while BCFtools/csq adds haplotype context. Differences should be explained by phase and transcript inputs.
Working checkpoint — functional validation: Align assembly and annotation before comparing outputs. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 25 of 36
25. Read performance claims in context
The primary study and official guide describe strong speed and memory performance on the benchmark used. Hardware, file size, version and options still shape real workloads.
Working checkpoint — haplotype-aware consequence: Benchmark the installed release on representative data. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 26 of 36
26. Audit one compound consequence
Choose a pair of phased variants and reconstruct both haplotype transcript sequences, codons and amino-acid outcomes manually.
Working checkpoint — phase information: Use this case as a permanent regression test. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 27 of 36
27. Use bounded clinical language
A haplotype-aware predicted consequence is more structurally faithful than isolated labels in relevant cases, but it is still an annotation.
Working checkpoint — compound variant: Require clinical and functional evidence for medical conclusions. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 28 of 36
28. Connect to Primary Science
Two transparent overlays can show how separate changes combine on one copy while changes on different copies do not.
Working checkpoint — GFF3 transcript: Use coloured paper strips, not patient data. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 29 of 36
29. Connect to Secondary Science
Chromosome pairs, alleles, transcription and translation meet in the haplotype model. Context determines how DNA changes alter a protein sequence.
Working checkpoint — BCSQ field: Separate inherited phase from predicted consequence. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 30 of 36
30. Connect to mathematics
Ordered pairs and combinations explain why the same unphased genotype can represent several haplotype arrangements.
Working checkpoint — functional validation: Enumerate possible arrangements before choosing one. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 31 of 36
31. Connect to computing
Efficient streaming over compressed BCF, indexed FASTA and formal GFF3 features turns a biological sequence problem into reproducible computation.
Working checkpoint — haplotype-aware consequence: Pin file indexes, schemas and software versions. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Section 32 of 36
32. Connect to pathways
Haplotype-aware annotation links genetics, algorithms and data engineering and can illuminate STEM education and career pathways.
Working checkpoint — phase information: Check current official courses for real pathway decisions. Read the numerical result together with sample-level plots, effect size and a sensitivity check capable of weakening the headline.
For future study, connect the idea to biology, mathematics and computing while checking current official pathway information separately.
Section 33 of 36
33. Did you know?
The official guide gives two memorable cases: adjacent variants can alter one codon, and a later indel can restore a frame disrupted earlier. One-row-at-a-time thinking can miss both.
Working checkpoint — compound variant: Ask whether nearby alleles were read together or separately. Preserve the complete tested family and every excluded feature; a polished shortlist is not a substitute for provenance.
For scientific writing, prefer “supports”, “is associated with” and “is predicted to” when stronger causal verbs are not justified.
Section 34 of 36
34. Write the result responsibly
A strong report names sample, phased genotype, contributing positions, transcript, combined consequence, phase source and validation status.
Working checkpoint — GFF3 transcript: Show the isolated predictions beside the combined result when instructive. Separate what the model estimates from the biological story proposed afterward, then name the extra evidence that story would require.
For students, the transferable habit is to ask: What was measured? Compared with what? What alternative explanation remains?
Section 35 of 36
35. Build a validation ladder
Verify alleles and phase, reconstruct transcripts, compare local and haplotype-aware outputs, inspect RNA and protein evidence, then test function where needed.
Working checkpoint — BCSQ field: Strengthen the weakest assumption first. Treat identifiers, denominators and factor levels as scientific quantities: a silent mismatch can reverse an otherwise correct calculation.
For parents, a calm checkpoint is to ask how many independent samples support the claim and which assumption matters most.
Section 36 of 36
36. Finish with a reproducible bundle
Deliver phased VCF or BCF, FASTA and indexes, Ensembl GFF3, BCFtools version and command, full BCSQ output, linked records, phase evidence and validation record.
Working checkpoint — functional validation: Make every compound consequence reconstructable on both haplotypes. Keep the independent sample, design, annotation and software version beside this decision so the evidence can be reconstructed.
For a classroom model, use invented counts and clearly label the activity as an analogy rather than a sequencing experiment.
Contents · Previous section · Continue to the Science Learning Hub