1 of 23

ga4gh.org

2 of 23

Sequence collections

how to uniquely identify and compare genomes

Andy Yates, Nathan Sheffield, Timothee Cezard

ga4gh.org

3 of 23

What’s a sequence collection ?

It’s a collection of sequences!

Like a reference genome

… Or a transcriptome

… Or a proteome

​

Sequence

Refget

hay6rlosvc4jdonnwy26fdk0

klavctwpjsccxoobfilld5oa67

SeqCol

ADR 2021-06-30

ga4gh.org

4 of 23

Use cases in brief

  • Cache and retrieve repetitive header information concerning sequences as required
  • Variation annotation and submission for archiving
  • A computable identifier for reference genomes for reproducibility
    • Refgenie reference genome asset identity
  • Assert compatibility between reference genomes
    • BED file comparison
    • Genome browser visualizations

ga4gh.org

5 of 23

Introduction to Sequence Collections

Project goal:

  • to standardize unique identifiers for collections of sequences
  • can be used to identify genomes, transcriptomes, or proteomes -- anything that can be represented as a collection of sequences

​

The project specifies:

  • an algorithm for computing sequence identifiers from collections
  • a lookup API to retrieve a collection given an identifier
  • a comparison API to assess compatibility of two collections

​

​

ga4gh.org

6 of 23

How do we represent a sequence collection ?

JSON object where each sequence collection attribute is a property of the JSON

​

Sequences as refget digests

Names used for these sequences

Length of the sequences

ADR 2021-06-30

ga4gh.org

7 of 23

How do we represent a sequence collection ?

JSON object where each sequence collection attribute is a property of the JSON

​

ADR 2021-06-30

ga4gh.org

8 of 23

How do we represent a sequence collection ?

JSON object where each sequence collection attribute is a property of the JSON

​

ADR 2021-06-30

ga4gh.org

9 of 23

Computing sequence identifiers

General algorithm

​

​

Normalisation: convert JSON into unique string representation

Using RFC8785: JSON Canonicalization Scheme (JCS)

​

​

Hashing: SHA512 truncated to the first 24 bits and converted to base64

seqcol

Normalisation

Hashing

digest

ADR 2023-01-12

ga4gh.org

10 of 23

Computing sequence identifiers

General algorithm

​

​

2

1

3

4

ga4gh.org

11 of 23

Computing sequence identifiers

Inherent/non-inherent attributes

  • Inherent attributes are included in the calculation of the identifier
  • Non-inherent attributes enables:
    • Storing additional metadata

​

​

​

​

    • Storing comparison helpers

sorted-name-length

ga4gh.org

12 of 23

Computing sequence identifiers

Inherent/non-inherent attributes

  • Inherent attributes are included in the calculation of the identifier
  • Non-inherent attributes enables:
    • Storing additional metadata

​

​

​

​

    • Storing comparison helpers

sorted-name-length

sorted-names-lengths attribute

ga4gh.org

13 of 23

API specification

ADR 2021-12-01

ga4gh.org

14 of 23

API specification (Lookup)

ga4gh.org

15 of 23

API specification (Comparison)

ADR 2022-06-15

ga4gh.org

16 of 23

API demonstration

https://seqcolapi.databio.org/static/links_ref.html

ADR 2022-06-15

ga4gh.org

17 of 23

Sequence Collections status and future

ADR specificies:

  • an algorithm for computing sequence identifiers from collections
  • a lookup API to retrieve a collection given an identifier
  • a comparison API to assess compatibility of two collections

​

Future work:

  • Finalise service-info
  • Write up specification
  • Implementations
  • Integration in tooling

​

​

ga4gh.org

18 of 23

Use case: Variation detection and annotation

Sequencing data

Variant call

​

Calculate

Sequence collection

Genome

chr1

chr2

…

VCF

##Seqcollection=<ID=oshqvxks73nsa>

Variant annotation

SeqCol

Server

Annotated VCF

##Seqcollection=<ID=oshqvxks73nsa>

Check compatibility of oshqvxks73nsa with supported annotation

Annotation

1

2

…

ga4gh.org

19 of 23

Use case: Variation Submission for Archiving

SeqCol

Server

Annotated VCF

##Seqcollection=<ID=oshqvxks73nsa>

Check compatibility of oshqvxks73nsa with supported genome

Genome

Annotated VCF

##Seqcollection=<ID=oshqvxks73nsa>

Validator

ga4gh.org

20 of 23

Use case: Tracking Refgenie asset reference identity

ga4gh.org

21 of 23

Use case: Coordinate system and visualisation

ga4gh.org

22 of 23

Use case: Caching header information for gains!

Retrieve/cache locally

(see CRAM)

Formatted header

ga4gh.org

23 of 23

Conclusion: get involved

Read and comment on the draft specification here:

​

View a demo implementation here: https://seqcolapi.databio.org

Software: https://github.com/refgenie/seqcol

Join our calls at:

  • Every other Wednesday at 2pm GMT / 10am ET
  • Reach out to reggan.thomas@ga4gh.org to get involved & invites

​

ga4gh.org