UG200 / UG200 Ultra — Ultima Genomics¶
Ultima's wafer-based single-end SBS platform at production scale — 162 Tb/wk ceiling and Q60 ppmSeq duplex-equivalent reads, optimized for cohort-scale exome and WGS.
At a glance¶
- Vendor / launch date / availability: Ultima Genomics; presented and shipping at AGBT 2026 per Vilella Substack Part 1.
- Chemistry / modality: Single-end sequencing-by-synthesis on circular silicon wafers using isothermal amplification; paired-end ("Solaris 2.0") on the 2026 roadmap per Vilella Substack Part 1.
- Throughput (PR): Up to 162 Tb/week maximum; 60 wafers/week throughput on UG200 Ultra; ~1,350 genome-equivalents/week (Ultra) per Vilella Substack Part 1 (vendor PR — no independent benchmark).
- Read length: Up to ~300 bp single-end; paired-end on the Solaris 2.0 roadmap per Vilella Substack Part 1.
- Accuracy (PR): Q60 ppmSeq duplex-equivalent per Vilella Substack Part 1 (vendor PR — no independent benchmark).
- List price / cost-per-genome: $850k (UG200 / single wafer) and $1.25M (UG200 Ultra) per Vilella Substack Part 1.
What's new vs prior generation¶
UG200 follows the UG100 wafer-sequencer, doubling effective wafer throughput and introducing the UG200 Ultra tier (multi-wafer, ~1,350 genomes/week). PpmSeq accuracy moves from ~Q40 duplex-equivalent on the prior generation to Q60 on UG200, narrowing the duplex-accuracy gap with PacBio HiFi and Roche SBX while preserving the single-end / cohort-scale economics that have been Ultima's pitch since launch. The Solaris 2.0 paired-end roadmap, if it ships, would close the historical gap vs Illumina/Element for variant callers tuned to PE inserts.
Headline benchmarks¶
- 162 Tb/wk ceiling and ~1,350 genome-equivalents/week on UG200 Ultra (vendor PR — no independent benchmark) — Vilella Substack Part 1.
- Q60 ppmSeq duplex-equivalent across the production wafer (vendor PR — no independent benchmark) — Vilella Substack Part 1.
Reception / competitive context¶
UG200 is the throughput ceiling of the AGBT 2026 short-read lineup; Element VITARI and Complete DNBSEQ-T7+ compete on price ($100 vs Ultima's higher capital cost) while Ultima retains the per-week genome-count advantage. Roche Axelios's $150 genome is an order of magnitude lower in instrument cost ($750k) but ships later (summer 2026). GenomeWeb's "battle royale" recap and the GEN AGBT recap both put Ultima in the "incumbent at cohort scale" position, with Element as the principal challenger.
Cross-links¶
- Downstream Bioconductor tools: short-read aligners + variant callers; methylKit for any duplex methylation work; cohort-scale workflows in [GBCC 2025 spatialLIBD-adjacent population-genomics talks].
- Competitive context in this AGBT lineup: Element VITARI (price challenger), Roche Axelios 1 (cost-per-genome floor), Complete DNBSEQ-T7+ ($100 genome twin).
- AACR 2026 tie-in: AACR posters tagged "Ultima" / "single-end" + cancer cohort scale are the natural first wave of UG200-era data.
Sources¶
- Vilella Substack Part 1 — Illumina announcements and Ultima — Feb 2026.
- GenomeWeb — "Battle Royale" recap — Feb 2026.
- GEN — AGBT 2026 Recap: NGS Big Bets and Spatial's Rising Momentum — Feb–Mar 2026.