eduKateSG · Why Science?
Pack an entire reference genome for fast retrieval—while keeping sequence identity, masking, unknown bases and provenance visible
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. A UCSC .2bit file stores multiple DNA sequences in a compact, randomly accessible binary representation. The official specification describes a 16-byte header, a name-to-offset index and sequence records that contain DNA length, blocks of unknown N bases, lower-case masking blocks and DNA packed at two bits per A, C, G or T base. UCSC provides faToTwoBit, twoBitInfo and twoBitToFa for conversion, inspection and extraction, including remote regional extraction. Compact access does not prove that the file contains the intended assembly, preserves the desired masking policy or matches a named reference release. 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: FASTA sequence records and reference data; FAI and random access to FASTA sequences; refget and globally identifiable reference sequences; Science Learning Hub; Education Hub; STEM Education; Careers by Subject Capability.
Primary and official sources checked on 12 October 2026: UCSC Genome Browser TwoBit Sequence Archives, checked 12 October 2026; UCSC Genome Browser .2bit format specification, checked 12 October 2026; UCSC Genome Browser REST API sequence endpoint documentation, 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, tools and recommendations can change, so preserve the dated documentation, exact source data, coordinate rules and software versions used.
A deeper reading route
Two bits are enough to distinguish four canonical DNA symbols. The format maps T, C, A and G to four binary codes and packs four bases into one byte. That efficiency is only part of the design. Unknown bases and lower-case masking cannot fit into the same four-symbol code, so the sequence record stores their positions and sizes as separate block lists that are restored during decoding.
Random access begins with the index. After the header, every sequence name points to the byte offset of its record. A reader can locate chr1 or another contig without decoding all earlier sequences. This resembles a library catalogue: the index accelerates retrieval, while the sequence record and its provenance still determine what is actually on the shelf.
The header supports portability. Its signature lets a reader detect whether multi-byte integers need byte swapping, the version tells a reader whether it understands the layout, and sequenceCount declares how many index entries follow. A familiar .2bit suffix is not evidence that these structural checks passed.
Masking carries scientific meaning. UCSC notes that lower-case input is considered masked in twoBit and can be ignored by gfServer when a mask option is used. Converting everything to upper case can be appropriate for one application and destructive for another. The choice should be declared, tested and separated from the underlying reference sequence identity.
Creation normally begins with FASTA and faToTwoBit. UCSC then demonstrates twoBitInfo for listing sequence sizes and twoBitToFa for extracting a sequence or coordinate range. A successful conversion proves that the bytes fit the format; it does not prove that FASTA headers, contig names, unknown regions or masking agree with the intended assembly release.
Reference identity needs more than a filename. Store the assembly provider, release, sequence names and lengths, input and output hashes, conversion command and any normalization of headers or case. A refget digest or equivalent per-sequence checksum can help distinguish two references that share a label but differ in bases.
Validation should exercise structure and meaning. Compare every sequence name and length, extract known ranges at starts and ends, count unknown and masked blocks, round-trip representative contigs to FASTA and compare decoded sequence under the declared case policy. Include empty mask lists, long N runs and a final partial byte in the fixture set.
For students, 2bit joins genetics and computing with a wonderfully tangible idea: four DNA letters fit into four binary states. The deeper scientific lesson is even better. Compression is trustworthy only when the decoder, index, reference identity and exceptional regions travel with the packed data.
Operational archives should test retrieval after migration, not merely compare file size. Host a versioned object, request several known ranges over the same transport path used by clients, verify returned bases and record range, object identity and software version. If an API or cache is involved, compare its result with direct twoBitToFa extraction so a stale service does not masquerade as the current reference.
Case normalization deserves its own release note. Converting lower case to upper case can preserve canonical base identity while deleting the mask state used by repeat-aware software. If two archives differ only in masking, whole-file hashes and decoded case differ even when uppercase sequence digests match. Publish both comparisons so users know whether the package changed biologically relevant sequence or only its annotation layer.
A classroom audit can reveal the whole design with a tiny eight-base sequence. Students encode canonical bases into two bytes, list one N block and one masked block, then reconstruct the original text. The exercise shows that compression depends on a codebook plus exception tables. It also encourages a valuable habit: never throw away the exceptions just because the common case is compact.
Long-term preservation benefits from a plain-text inventory. Keep one row per sequence with name, length, uppercase digest, case-sensitive digest where appropriate and N or mask-block counts. The inventory makes a binary archive inspectable without replacing it and lets future tools confirm logical equivalence even if repackaging changes the exact bytes.
| Stage | Question | Evidence | Responsible output |
|---|---|---|---|
| Identify | Which reference sequences are intended? | Assembly release, names, sizes and sequence hashes | A defined reference set |
| Pack | How are bases and exceptional regions represented? | Two-bit DNA plus N and mask blocks | A compact sequence archive |
| Retrieve | How is one contig or range located? | Name index, offsets and twoBitToFa or API query | A bounded sequence result |
| Verify | Did conversion preserve the intended reference? | Round trips, checksums, ranges and masking tests | A reproducible reference object |
Inside this guide
1–12 · Foundations and design
- 1. Start with the reference question
- 2. Recognise a binary sequence archive
- 3. Understand two bits per canonical base
- 4. Keep unknown bases separate
- 5. Keep masking separate
- 6. Respect the documented size boundary
- 7. Read the 16-byte header
- 8. Check the signature
- 9. Check the version
- 10. Count index entries
- 11. Read variable-length names
- 12. Follow offsets cautiously
13–24 · Models and estimation
- 13. Read dnaSize
- 14. Read N-block cardinality
- 15. Read mask-block cardinality
- 16. Keep reserved fields zero
- 17. Decode packed order correctly
- 18. Handle a final partial byte
- 19. Create from controlled FASTA
- 20. Inspect with twoBitInfo
- 21. Extract with twoBitToFa
- 22. Use remote access deliberately
- 23. Separate coordinates from sequence identity
- 24. Preserve case policy
25–36 · Diagnostics and interpretation
- 25. Validate a round trip
- 26. Hash at file and sequence levels
- 27. Protect human genomic references
- 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 reference note
- 35. Build a validation ladder
- 36. Finish with a reproducible bundle
Section 1 of 36
1. Start with the reference question
A 2bit archive should represent a named set of reference sequences for a defined workflow. Storage format cannot choose the correct assembly.
Working checkpoint — 2bit file: Write provider, assembly release, contig policy and intended use first. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Section 2 of 36
2. Recognise a binary sequence archive
UCSC describes 2bit as a compact randomly accessible format for multiple DNA sequences. It is not a read file, alignment file or annotation track.
Working checkpoint — packed DNA: Identify the scientific object before choosing the format. Read every extraction beside its sequence name, half-open range, reference release and decoded case.
For future study, connect genetics to binary systems and archives while checking pathways separately.
Section 3 of 36
3. Understand two bits per canonical base
Four binary states encode T, C, A and G. Packing four bases per byte reduces repeated textual storage.
Working checkpoint — sequence index: Decode a short known sequence by hand. Preserve unknown and masked blocks; compact canonical bases are not the whole sequence model.
For scientific writing, report identity, masking, conversion and validation.
Section 4 of 36
4. Keep unknown bases separate
N bases represent unknown sequence and are restored from nBlockStarts and nBlockSizes rather than the packed canonical codes.
Working checkpoint — N block: Compare N blocks with the source FASTA. Separate binary integrity, logical sequence identity and biological interpretation.
For students, ask how four letters become four binary codes and where exceptions are stored.
Section 5 of 36
5. Keep masking separate
Lower-case masked regions are represented with mask-block arrays and restored independently from base identity.
Working checkpoint — mask block: State whether case was preserved, removed or created. Treat offsets, names, sizes, Ns and masking as scientific inputs.
For parents, ask whether the compact file still has a named source and release.
Section 6 of 36
6. Respect the documented size boundary
The UCSC FAQ describes the format as holding multiple sequences up to four gigabases in total. Workflows near limits need explicit testing.
Working checkpoint — reference identity: Measure the input and reject unsupported scale assumptions. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Section 7 of 36
7. Read the 16-byte header
The header contains signature, version, sequenceCount and a reserved field, each currently 32 bits.
Working checkpoint — 2bit file: Validate all four values before following offsets. Read every extraction beside its sequence name, half-open range, reference release and decoded case.
For future study, connect genetics to binary systems and archives while checking pathways separately.
Section 8 of 36
8. Check the signature
The specified signature detects the format and indicates when byte swapping is required for multi-byte entities.
Working checkpoint — packed DNA: Reject a file whose signature fails both byte orders. Preserve unknown and masked blocks; compact canonical bases are not the whole sequence model.
For scientific writing, report identity, masking, conversion and validation.
Section 9 of 36
9. Check the version
The current documented version is zero, and readers should stop on a higher unsupported version.
Working checkpoint — sequence index: Fail closed instead of guessing a future layout. Separate binary integrity, logical sequence identity and biological interpretation.
For students, ask how four letters become four binary codes and where exceptions are stored.
Section 10 of 36
10. Count index entries
sequenceCount determines how many name-and-offset entries follow the header. Truncation can make the index incomplete.
Working checkpoint — N block: Reconcile declared and decoded sequence counts. Treat offsets, names, sizes, Ns and masking as scientific inputs.
For parents, ask whether the compact file still has a named source and release.
Section 11 of 36
11. Read variable-length names
Each index entry stores one-byte name length followed by the sequence name. Names connect the archive to chromosome conventions.
Working checkpoint — mask block: Compare names byte for byte with the manifest. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Section 12 of 36
12. Follow offsets cautiously
Each index offset points from the file start to one sequence record and need not align to a four-byte boundary.
Working checkpoint — reference identity: Check that every offset is inside the file and non-overlapping as expected. Read every extraction beside its sequence name, half-open range, reference release and decoded case.
For future study, connect genetics to binary systems and archives while checking pathways separately.
Section 13 of 36
13. Read dnaSize
The sequence record declares the number of DNA bases. Extraction bounds and packed byte counts depend on it.
Working checkpoint — 2bit file: Compare dnaSize with twoBitInfo and source FASTA length. Preserve unknown and masked blocks; compact canonical bases are not the whole sequence model.
For scientific writing, report identity, masking, conversion and validation.
Section 14 of 36
14. Read N-block cardinality
nBlockCount controls the lengths of the N-start and N-size arrays. Malformed counts can misalign every later field.
Working checkpoint — packed DNA: Validate array lengths and interval bounds. Separate binary integrity, logical sequence identity and biological interpretation.
For students, ask how four letters become four binary codes and where exceptions are stored.
Section 15 of 36
15. Read mask-block cardinality
maskBlockCount similarly controls masked-region arrays and may be zero. Masking is metadata with analytical consequences.
Working checkpoint — sequence index: Test both masked and unmasked fixtures. Treat offsets, names, sizes, Ns and masking as scientific inputs.
For parents, ask whether the compact file still has a named source and release.
Section 16 of 36
16. Keep reserved fields zero
The documented reserved fields are currently zero. Non-zero values may indicate corruption or an unsupported extension.
Working checkpoint — N block: Report rather than silently ignoring an unexpected value. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Section 17 of 36
17. Decode packed order correctly
The first base occupies the most significant two bits of a byte and later bases follow toward the least significant bits.
Working checkpoint — mask block: Test TCAG against the documented binary example. Read every extraction beside its sequence name, half-open range, reference release and decoded case.
For future study, connect genetics to binary systems and archives while checking pathways separately.
Section 18 of 36
18. Handle a final partial byte
A sequence length need not be divisible by four, so the final byte can contain unused low-order states.
Working checkpoint — reference identity: Compare the exact decoded length, not all possible bit slots. Preserve unknown and masked blocks; compact canonical bases are not the whole sequence model.
For scientific writing, report identity, masking, conversion and validation.
Section 19 of 36
19. Create from controlled FASTA
faToTwoBit converts prepared FASTA to 2bit. Header normalization, whitespace and case choices should be fixed before conversion.
Working checkpoint — 2bit file: Archive the exact FASTA hash and command. Separate binary integrity, logical sequence identity and biological interpretation.
For students, ask how four letters become four binary codes and where exceptions are stored.
Section 20 of 36
20. Inspect with twoBitInfo
UCSC demonstrates twoBitInfo for listing sequences and sizes and for creating a chrom.sizes file.
Working checkpoint — packed DNA: Save and compare the complete listing. Treat offsets, names, sizes, Ns and masking as scientific inputs.
For parents, ask whether the compact file still has a named source and release.
Section 21 of 36
21. Extract with twoBitToFa
twoBitToFa can retrieve full sequences or coordinate ranges from local or remote archives.
Working checkpoint — sequence index: Test full-contig and bounded regional extraction. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Section 22 of 36
22. Use remote access deliberately
UCSC tools can operate on a .2bit URL and request a range, making stable hosting part of reproducibility.
Working checkpoint — N block: Record URL, object size, hash and access time. Read every extraction beside its sequence name, half-open range, reference release and decoded case.
For future study, connect genetics to binary systems and archives while checking pathways separately.
Section 23 of 36
23. Separate coordinates from sequence identity
A correct range against the wrong assembly returns plausible but incorrect bases. Chromosome labels alone are insufficient.
Working checkpoint — mask block: Bind every query to the exact reference manifest. Preserve unknown and masked blocks; compact canonical bases are not the whole sequence model.
For scientific writing, report identity, masking, conversion and validation.
Section 24 of 36
24. Preserve case policy
Lower-case bases may mark repeats or another masking rule. Some applications ignore masked sequence when configured to do so.
Working checkpoint — reference identity: Explain what masking means for this specific release. Separate binary integrity, logical sequence identity and biological interpretation.
For students, ask how four letters become four binary codes and where exceptions are stored.
Section 25 of 36
25. Validate a round trip
Convert representative sequence content to 2bit and back, then compare bases, Ns, case and headers under declared rules.
Working checkpoint — 2bit file: Distinguish intentional header changes from data loss. Treat offsets, names, sizes, Ns and masking as scientific inputs.
For parents, ask whether the compact file still has a named source and release.
Section 26 of 36
26. Hash at file and sequence levels
A whole-file hash identifies exact bytes, while per-sequence hashes help compare logical reference content across packaging changes.
Working checkpoint — packed DNA: Store both kinds when long-term identity matters. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Section 27 of 36
27. Protect human genomic references
Reference assemblies are generally public, but custom or individual sequences can be sensitive even in compact form.
Working checkpoint — sequence index: Apply governance based on sequence content, not extension. Read every extraction beside its sequence name, half-open range, reference release and decoded case.
For future study, connect genetics to binary systems and archives while checking pathways separately.
Section 28 of 36
28. Connect to Primary Science
Four coloured cards can model four DNA bases and paired cards can model two-bit codes without using real genomic data.
Working checkpoint — N block: Ask pupils to separate code from biological meaning. Preserve unknown and masked blocks; compact canonical bases are not the whole sequence model.
For scientific writing, report identity, masking, conversion and validation.
Section 29 of 36
29. Connect to Secondary Science
Genetics lessons can distinguish a reference model from an individual genome and an unknown N from a confirmed base.
Working checkpoint — mask block: Use precise language about what the archive represents. Separate binary integrity, logical sequence identity and biological interpretation.
For students, ask how four letters become four binary codes and where exceptions are stored.
Section 30 of 36
30. Connect to mathematics
Binary numbers, powers of two, offsets and intervals make 2bit a concrete discrete-mathematics example.
Working checkpoint — reference identity: Calculate how many canonical bases fit in one byte. Treat offsets, names, sizes, Ns and masking as scientific inputs.
For parents, ask whether the compact file still has a named source and release.
Section 31 of 36
31. Connect to computing
Indexes, byte order, variable-length records, binary parsing and random access show how data structures support science.
Working checkpoint — 2bit file: Validate lengths before allocating or seeking. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Section 32 of 36
32. Connect to pathways
Reference data work links genomics, software engineering, archives and research infrastructure and can illuminate STEM pathways.
Working checkpoint — packed DNA: Check current official qualifications separately. Read every extraction beside its sequence name, half-open range, reference release and decoded case.
For future study, connect genetics to binary systems and archives while checking pathways separately.
Section 33 of 36
33. Did you know?
The four canonical DNA bases fit into two bits, but Ns and lower-case masking are restored from separate interval lists.
Working checkpoint — sequence index: Explain the exception tables as part of the format. Preserve unknown and masked blocks; compact canonical bases are not the whole sequence model.
For scientific writing, report identity, masking, conversion and validation.
Section 34 of 36
34. Write a reader-ready reference note
Name provider, release, source FASTA, case policy, tools, names, sizes, hashes, hosting and validation.
Working checkpoint — N block: Keep it beside every downstream alignment or track. Separate binary integrity, logical sequence identity and biological interpretation.
For students, ask how four letters become four binary codes and where exceptions are stored.
Section 35 of 36
35. Build a validation ladder
Check header, index, offsets, record lengths, N blocks, mask blocks, packed bases, sequence hashes and remote queries.
Working checkpoint — mask block: Stop at the first unexplained difference. Treat offsets, names, sizes, Ns and masking as scientific inputs.
For parents, ask whether the compact file still has a named source and release.
Section 36 of 36
36. Finish with a reproducible bundle
Deliver source hash, 2bit hash, manifest, conversion command, twoBitInfo output, extraction fixtures and validation logs.
Working checkpoint — reference identity: Make every retrieved base traceable to one reference release. Keep source FASTA, assembly manifest, case policy, conversion command, twoBit hash and sequence checksums together.
For a classroom model, use invented DNA letters and no personal data.
Contents · Previous section · Continue to the Science Learning Hub