Why current spatial omics service workflows fail in practice
I once processed a frozen human cortex block in Shenzhen (June 2021) and saw usable UMI yield drop by 40% after seven days of cold storage — what is the measurable loss other labs accept as normal? Early in that run I consulted the stereo-seq inventor notes and realized the gap was not sequencing per se but sample handling upstream. In my view, spatial omics service offerings too often promise turnkey spatial transcriptomics results while glossing over fixation, tissue embedding, and barcode array alignment issues — this is where studies silently fail (and clients pay for retries).
I’ve run stereo-seq on glass slides and on fresh-frozen tissue, and I can tell you the common fixes vendors recommend are blunt instruments. They raise sequencing depth and call it solved. They ignore how poor tissue fixation and inconsistent RNA capture inflate background noise and collapse meaningful signal. I remember a January 2022 pilot where a lab increased sequencing depth by 30% and still lost cell-type resolution because the barcode array had a 10–15 µm misregistration — no kidding. Those are the traditional solution flaws that a service-level SLA rarely admits. The result: wasted reads, inflated cost-per-sample, and delayed publication timelines. — Here’s the hard part: most customers ask for throughput, not for a breakdown of per-step failure modes, and that mismatch drives repeated problems.
Transitioning to next steps requires a straightforward audit of where reads vanish.
Forward-looking fixes and how to evaluate next-gen spatial omics service partners
We must shift from shotgun fixes to controlled process metrics. I now require three checkpoints before any service engagement: validated tissue fixation protocol, in-run barcode array QC, and a clear sequencing depth plan tied to target spatial resolution. When I restructured a 2023 project for a mid-size biotech, I specified 2 µm target resolution, a preliminary barcode registration run, and a staged sequencing depth increase (20M → 35M reads) — that exposed a tissue-prep artifact before we burned budget. That stepwise approach stems directly from lessons I learned using the stereo-seq inventor method notes and from hands-on runs in two different core facilities.
What’s Next?
Technically, improvements are incremental: better RNA preservation chemistry, tighter barcode array calibration, and smarter UMI deduplication algorithms. I advocate for real pre-run QC: simple fluorescence checks for RNA integrity, spot-checks for barcode registration, and pilot sequencing at low depth to test library complexity. This is practical. It works. — Two brief interruptions: verify buffers; re-check slide orientation. These small acts save months.
To choose a spatial omics service partner I recommend three concrete evaluation metrics: (1) Pre-run QC coverage — percent of runs with documented tissue integrity checks; (2) Recovery efficiency — average usable UMI per mm² at declared resolution; (3) Reproducibility window — fraction of replicate sections that meet caller thresholds within one run. Use those metrics to compare proposals, and demand raw QC files (fastq/QC reports) alongside final matrices. I speak from direct experience: a supplier who could provide those numbers for a pilot in March 2022 earned my team’s repeat business.
Final note: I want services that admit their failure modes and measure them. That honesty—paired with concrete QC metrics—changes outcomes. For practical partnerships and validated implementation work, check stomics.