Representing specific flow cells for DNA sequencers

[cmungall]

cc @turbomam

If we need to represent Illumina NovaSeq 6000 S4, I assume we would create a subclass of

id: OBI:0400043
name: flow cell
def: "Aparatus in the fluidic subsystem where the sheath and sample meet. Can be one of several types; jet-in-air, quartz cuvette, or a hybrid of the two. The sample flows through the center of a fluid column of sheath fluid in the flow cell." []
synonym: "flow_cell" EXACT []
relationship: RO:0000085 OBI:0000370 ! contain function
is_a: OBI:0000967 ! container

and then make a part_of link to the sequencer, eg.

id: OBI:0002630
name: Illumina NovaSeq 6000
def: "A DNA sequencer which is manufactured by the Illumina corporation, with two flow cells and an output of up to 6000 Gb (32-40 B reads per run). The sequencer utilizes synthesis technology and patterned flow cells to optimize throughput and even spacing of sequencing clusters." []
synonym: "NovaSeq 6000" EXACT []
is_a: OBI:0400103 ! DNA sequencer

Would this be a new robot template? E.g. something like:

Sequencers	flow cell	description
Illumina HiSeq 2500	High Output flow cell	Higher throughput but longer run times than Rapid Run mode
Illumina HiSeq 2500	Rapid Run flow cell	Faster run times but lower output than High Output mode
Illumina HiSeq 3000	patterned flow cell	Similar to HiSeq 4000 but with lower throughput
Illumina HiSeq 4000	patterned flow cell	High-output flow cell for various applications
Illumina HiSeq X	patterned flow cell	High-throughput flow cell optimized for human whole-genome sequencing
Illumina iSeq 100	iSeq 100 i1 flow cell	Single flow cell type for iSeq 100
Illumina MiSeq	v1	Lowest output flow cell for MiSeq
Illumina MiSeq	v2	Mid-range output flow cell for MiSeq
Illumina MiSeq	v3	Highest output flow cell for MiSeq
Illumina NextSeq 550	High Output flow cell	Higher capacity flow cell for NextSeq 550
Illumina NextSeq 550	Mid Output flow cell	Lower capacity flow cell for NextSeq 550
Illumina NovaSeq 6000	S1	Lower output flow cell for NovaSeq 6000
Illumina NovaSeq 6000	S2	Mid-range output flow cell for NovaSeq 6000
Illumina NovaSeq 6000	S4	Highest output flow cell for NovaSeq 6000, up to 10B reads
Illumina NovaSeq 6000	SP	Lowest output flow cell for NovaSeq 6000, fastest run times
Ion Torrent PGM	Ion 314 chip	Lowest throughput chip for PGM
Ion Torrent PGM	Ion 316 chip	Mid-range throughput chip for PGM
Ion Torrent PGM	Ion 318 chip	Highest throughput chip for PGM
Ion Torrent Proton	Ion PI chip	High-throughput chip for Proton
Ion Torrent S5	Ion 520 chip	Low to mid-range throughput chip for S5
Ion Torrent S5	Ion 530 chip	Mid to high-range throughput chip for S5
Ion Torrent S5	Ion 540 chip	Highest throughput chip for S5
Oxford Nanopore GridION	R9.4.1 flow cell	Standard flow cell for GridION (same as MinION)
Oxford Nanopore MinION	R10.3 flow cell	Newer flow cell with improved accuracy for MinION
Oxford Nanopore MinION	R9.4.1 flow cell	Standard flow cell for MinION
Oxford Nanopore PromethION	R9.4.1 flow cell	High-capacity flow cell for PromethION
Pacific Biosciences Sequel	SMRT Cell 1M	Standard flow cell for original Sequel system
Pacific Biosciences Sequel II	SMRT Cell 8M	High-capacity flow cell for Sequel II system

[turbomam]

I think this has potential, but OBI:0400043 is a cytometry flow cell, not a sequencer flow cell. I'm pretty sure there's no such thing as a jet-in-air sequencer flow cell.

[turbomam]

I like this but it will result in a lot of pre-composed classes over time. Who's going to maintain them, or keep the set complete?

[bpeters42]

I think we should share with Chris our issues we have had in trying to maintain definitions for instruments. We initially wanted to follow an approach like Chris has here, but the problems are that: - much of the information on instruments come from marketing materials, which are full of 'fast', 'high throughput', 'massively parallel' etc. statements that don't age well at all. - The same machine may be sold under different names for marketing reasons or due to company mergers - We have found that most users just want to refer to what sequencer they used, and really just want a label and identifier; that is what we have committed to maintaining.

…

On Thu, Sep 12, 2024 at 11:35 AM Mark Andrew Miller < ***@***.***> wrote: I like this but it will result in a lot of pre-composed classes over time. Who's going to maintain them, or keep the set complete? — Reply to this email directly, view it on GitHub <#1820 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADJX2ITKYXLZ33DBRON7T6LZWGYCNAVCNFSM6AAAAABOBXSD42VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNBWGYZDKNBVGA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Bjoern Peters Professor La Jolla Institute for Immunology 9420 Athena Circle La Jolla, CA 92037, USA Tel: 858/752-6914 Fax: 858/752-6987 http://www.liai.org/pages/faculty-peters

[turbomam]

OBI's root DNA Sequencer class can be found here in OLS

[aclum]

Mark and I talked about this a bit offline, I care less about the flowcell IDs since it impacts throughput but not analysis. What I would like to see is

1. Illumina NovaSeq X added
2. Groupings by series (ex Illumina HiSeq, Illumina NovaSeq, etc), this makes the ontology easier to read and we have cases where we don't have the model number (ex we know it is illumina NextSeq but not if it is 500 or 550.

[bpeters42]

Both of those requests are straightforward to implement, and make perfect sense. I am hoping @turbomam can compile all the manufacturer / series you will need, and get it into OBI.

[sebastianduesing]

Illumina NovaSeq X was added recently in PR 1804, which will go live in the next OBI release.

[mgiglio99]

I second the motion for grouping terms for the sequencer series. I've also run into situations of not knowing which exact member of the series is in play.

[mgiglio99]

Discussed on Sept 30 call - general agreement that grouping terms for the sequencer series is a good idea.
Mark is going to provide a table with info on related terms from different sources.
He will propose grouping terms along the lines of this example: 'Illumina HiSeq Series sequencer' .

[turbomam]

I have started two two new issues and propose closing this issue about flow cells

1. Add is_manufactured_by Illumina to all Illumina brand DNA sequencers #1824
2. Add grouping classes for series of DNA sequencers #1825
   • following the summary @mgiglio99 gave above