Transcriptomic Insights

Attending the 5th Annual Nordic Digital Science & Innovation Day today in Copenhagen.   AI, digital infrastructure, data-driven laboratory practices — when working with transcriptomics, the value of your data is decided long before it hits the pipeline. Having worked across the full process myself, I know that high quality data starts with high quality lab work.   As AI transforms how we analyze and interpret data, the fundamentals of good lab work matter more than ever.   That end-to-end perspective is what drives how I work, and why I care about getting it right from the very first step.   Looking forward to the conversations today. #NordicDigitalScienceDay #Transcriptomics #SingleCell #Bioinformatics #LifeScience #DataDriven #AI

Introducing StrataMap Spatial, a powerful end-to-end whole-transcriptome solution with unmatched breadth of coverage and resolution    StrataMap Spatial expands researchers’ ability to understand the structure and function of tissues, study tumor microenvironments, and identify novel drug targets for precision medicine.     Learn how StrataMap Spatial is redefining what researchers can detect 🔗 : www.illumina.com/stratamap #EACR2026

A must read for anyone working with spatial transcriptomics data! 📖 New review from the Plummer lab covering in detail current spatial transcriptomics analysis approaches and tools. The spatial field has evolved enormously in recent years. Commercial technologies from players like Bruker and 10x Genomics continue to push resolution, throughput and sensitivity beyond comprehension. Consequently, the field faces a data challenge – how do we transform matrices, coordinates and pixels into meaningful biology? The current review beautifully dissects the general strategies for tackling this data – from initial steps like QC, segmentation and normalization on to downstream approaches for resolving cellular signaling, interactions and neighborhoods. It also provides an overview of data structures - if you're new in spatial analysis, this is an excellent place to start! Importantly, it also highlights current challenges like standardization and reproducibility. Many workflows are adapted from single cell RNA-seq pipelines rather than tailored for spatial data, cross-platform benchmarking is rare (I'm personally scouting for a detailed head-to-head comparison of Atera, Xenium and CosMx!), and well-annotated spatial reference atlases remain fragmented and tissue-specific. The terabyte-sized data doesn't make it any easier. Handling data of this size challenges approaches and infrastructure. The field is moving at breakneck speed but data analysis is still in its infancy. Going to be interesting to see how spatial analysis will uncover biological secrets. Highly recommend giving it a read (link in comments).

About ten years ago, I was working in the R&D department at Exiqon — developing NGS-based methods to profile microRNAs. It was exciting work, and the biology was compelling. But back then, getting non-coding RNA profiling to work robustly, let alone at scale, was genuinely hard. Exiqon is long gone now (acquired by Qiagen in 2016), but the biology hasn't gone anywhere. And I've been waiting to see this field properly enter the single-cell and spatial era ever since. The dominant approach in single-cell and spatial transcriptomics — 3'-end polyA capture — has been a workhorse for good reason: scalable, robust, and cost-effective. But it has a systematic blind spot. miRNAs, tRNAs, histone RNAs, many lncRNAs, and circular RNAs are largely excluded from most datasets today. A recent preprint from Isakova et al. (bioRxiv, 2025) is a meaningful step forward — demonstrating scalable total RNA profiling on the 10x Genomics Chromium platform, capturing non-coding RNA programs at single-cell resolution that simply don't appear in standard 3'-end data. This came to mind again at the recent single-cell symposium in Aarhus, where it was clear the field is ready to ask a bigger question: are we finally ready to look at the full transcriptome? And to be precise — even this new approach still has limits. Isoform resolution requires full-length transcript sequencing, and circular RNAs, which lack free ends entirely, remain out of reach. Emerging long read sequencing & random priming-based methods are beginning to address this — but that’s a conversation for another day. At Transcriptomic Insights, this is exactly the kind of methodological shift we watch closely — so we can help our partners design studies that capture the right biology, not just the convenient biology. #SingleCell #SpatialTranscriptomics #NonCodingRNA #microRNA #Transcriptomics #RNAseq #TranscriptomicInsights

On my way home from the 6th Danish Single Cell Symposium hosted by The Danish Single Cell Examination Platform. I’ve attended 5 out of the 6 meetings over the years — but this was Mette’s first. It’s amazing to see how rapidly the field continues to evolve from year to year. Also great to catch up with colleagues across the field and get a first look at many of the newest products and technologies presented by the vendors. A few personal takeaways: • Spatial transcriptomics increasingly feels like the center of gravity, while single-cell approaches are becoming a powerful complementary layer rather than a competing one • Are we approaching the limits of 3′ sequencing? There seems to be growing interest in moving beyond gene counts alone and bringing isoforms and non-coding biology into single-cell workflows • Multi-omics remains a major focus area — but increasingly it feels like the field is moving from promise to application. Especially exciting to see how single-cell proteomics continues to gain traction • Competition across platforms has never been stronger — across sequencing, single-cell, and spatial technologies • But perhaps the biggest bottleneck is no longer generating data — it’s understanding it. The volume and complexity of data being produced is accelerating rapidly, while analysis and biological interpretation still seem to lag behind the technologies themselves Thanks to organizers, speakers and poster-presenters for a great meeting! See you at the 7th meeting! #SingleCell #SpatialTranscriptomics #Transcriptomics

🚀 𝐅𝐢𝐫𝐬𝐭 𝐈𝐧𝐬𝐢𝐠𝐡𝐭𝐬 𝐢𝐧𝐭𝐨 𝐀𝐭𝐞𝐫𝐚 𝐒𝐩𝐚𝐭𝐢𝐚𝐥 𝐓𝐫𝐚𝐧𝐬𝐜𝐫𝐢𝐩𝐭𝐨𝐦𝐢𝐜𝐬 𝐃𝐚𝐭𝐚 Mariia Bilous and I recently had the opportunity to perform a preliminary analysis of the new Atera data released by 10x Genomics, focusing on the breast dataset. The attached figures summarize some of our initial observations. Overall, #Atera delivers sensitivity comparable to Chromium, consistent with 10x reports, both in terms of the number of genes detected and per-gene sensitivity. At the same time, the Xenium targeted panel profiled on the same tissue still shows higher sensitivity for the subset of commonly measured genes. Using Chromium as a reference for annotation, Atera and Xenium produced highly consistent cell-type compositions. Applying RCTD deconvolution in doublet mode, we observed reduced secondary cell-type signal in Atera relative to Xenium — likely reflecting lower transcript spillover — although some contamination remains. This improved resolution is also visible in the UMAP embeddings. We additionally tested our SPLIT method (Bilous et al., Nature Methods) on the Atera data and found that it further improves signal purity and cellular resolution in just a couple of minutes. Overall, I’m very impressed by the quality and resolution of the Atera data and particularly excited about the upcoming Atera 1K panels, which may offer an excellent balance between sensitivity and transcriptome breadth for clinical and translational applications. 𝐖𝐨𝐮𝐥𝐝 𝐈 𝐰𝐚𝐧𝐭 𝐚𝐧 𝐀𝐭𝐞𝐫𝐚 𝐢𝐧 𝐭𝐡𝐞 𝐥𝐚𝐛? 𝐀𝐛𝐬𝐨𝐥𝐮𝐭𝐞𝐥𝐲. That said, Xenium still has an important role to play, particularly when maximal sensitivity for targeted panels is required. Our two instruments are still running 24/7. 𝐒𝐡𝐨𝐮𝐥𝐝 𝐒𝐏𝐋𝐈𝐓 𝐛𝐞 𝐚𝐩𝐩𝐥𝐢𝐞𝐝 𝐭𝐨 𝐀𝐭𝐞𝐫𝐚 𝐝𝐚𝐭𝐚? 𝐀𝐛𝐬𝐨𝐥𝐮𝐭𝐞𝐥𝐲. It runs in just a couple of minutes and further improves signal purity, resolution, and overall data quality. We’ll share additional insights as we continue exploring the data and performing further analyses. In the meantime, if you want to learn more about SPLIT, please read our paper: https://lnkd.in/eGWAy4cX #SpatialTranscriptomics #SingleCell #ComputationalBiology #AIforBiology #CancerResearch

Interesting read 📖 in Nature Communications this week: Renoir, from the Zafar lab (IIT Kanpur) and Ankur Sharma's group. (Paper link in comments). It's a new framework for interrogating cell-to-cell communication in a niche-dependent manner based on spatial transcriptomics data. Unlike most CCC tools, which stop at ligand-receptor co-expression, Renoir factors in niche cell-type composition, receptor expression in the receiver cell, and downstream target gene activity. That matters, because cell-cell communication is anything but uniform. The same ligand often triggers very cellular responses depending on which cell type receives it, which receptor variant that cell expresses, and which other cells share the local microenvironment. A signal that drives proliferation in one niche can drive immunosuppression in another – context isn't a detail, it's the whole point. Tools like this really emphasize the value of spatial transcriptomics – the ability to map gene expression at high resolution while preserving spatial context. They open a world of possibilities with implications for everything from target discovery to personalized medicine. #SpatialTranscriptomics #CellCellCommunication