The features that make headlines are rarely the ones that decide whether a result holds up. That's usually settled earlier and quieter: in how carefully data is checked, how flexibly the platform handles a real-world study design, and how faithfully the path from raw data to figure is recorded. This release is a batch of exactly that kind of work.
Assays that fit your study, not the other way around
Real experiments don't always match a textbook template. nCounter gene expression experiments now work with one to three housekeeping genes, instead of requiring exactly three, so studies that were previously awkward to run just work. And single-nuclei RNA-seq (snRNA-seq) FASTQ experiments are now supported with their own dedicated summary parameters, reflecting how common frozen-tissue and hard-to-dissociate samples have become.
Catch problems before they cost you
Sample swaps and contamination are among the most expensive mistakes in genomics, precisely because they hide until late. Rosalind now runs biological QC for Kallisto-processed RNA-seq, helping surface those issues early, while you can still do something about them. You can also indicate whether an RNA-seq processed-counts experiment contains raw or normalized data right on the summary page, so downstream steps make the right assumptions.
Provenance, because reproducibility is a feature
The reproducibility conversation in life sciences has finally shifted from hand-wringing to tooling, and provenance is at the center of it. The new Publication module tracks figure exports and analysis provenance as you prepare a manuscript, so the story of how a figure was made travels with the figure. Alongside it, peak overlap track PNG downloads now render correctly at any screen resolution, and single sign-on registration is smoother for institutional teams.
None of this is flashy. But it's the substrate everything else stands on, and getting it right is how a platform earns the trust to sit in the middle of your science. It's also the groundwork for everything ahead: the more disciplined and well-provenanced your data is today, the more you'll be able to trust the smarter, more automated analysis we all know is coming. Rigor first, intelligence built on top of it. That order matters, and we're building in that order on purpose.
.png)
