scRNA-seq cell classification sounds straightforward until you actually try it. Sparse data. Thousands of genes. Millions of possible cell relationships. Most methods treat each cell as isolated. A new framework from Nanjing University flips that. 🧵 doi.org/10.1049/syb2... #ComputationalBiology

scGMB: A scRNA‐seq Cell Classification Method Combining GCN and Mamba

A single-cell RNA sequencing data classification method called scGMB is proposed. The method captures the topological relationships between cells.

doi.org

1/ Everyone's talking about AI in biology. But what does it actually look like when you apply Transformer models to scRNA-seq? A new survey in Briefings in Bioinformatics maps the whole landscape. Here's what's worth knowing. 🧵 doi.org/10.1093/bib/... #SingleCell #scRNAseq #Bioinformatics

Transformers for single-cell RNA sequencing: a survey

Abstract. Transformers have demonstrated remarkable success in the field of deep learning, attracting significant attention from researchers and driving in

doi.org

1/ The average bioinformatics tool is obsolete before most labs finish validating it. This is not an exaggeration, and it has real consequences for the science being produced. 🧵 #Bioinformatics #Genomics

scRNA-seq tells you what genes each cell is expressing. What it doesn't tell you is how those cells are related SCITE-RNA (Zimmermann, Sun, Hård et al., Genome Biology 2026) reconstructs tumor cell phylogenies directly from RNA sequencing data 🧵 doi.org/10.1186/s130... #scRNAseq #Bioinformatics

Targeted RNA-seq panels only find what they are designed to look for. A new study of 301 acute leukemia patients asks what whole RNA-seq finds instead. Kim, Park, Min et al. | Cells 2026 🧵 www.mdpi.com/2073-4409/15... #RNAseq #Bioinformatics #Leukemia #Genomics

Anti-TNF therapy fails roughly half of IBD patients. A new Gastroenterology paper identifies a molecular subgroup who may be systematically less likely to respond, and points to exactly why. 🧬 doi.org/10.1053/j.ga... #Trinobia #Bioinformatics #IBD #CrohnsDisease

Most scRNA-seq QC still comes down to three metrics: mt%, gene count, UMIs. The problem: cardiomyocytes, neurons, malignant cells, and erythroid cells can fail those filters simply because that is their biology. scQCenrich (Commun Biol 2026) was built to fix this. #scRNAseq #SingleCell

Preserving cells before scRNA-seq sounds straightforward. But which method holds up across labs, technicians, and shipping? A new multisite ABRF study tested 10x FLEX, Honeycomb HIVE, and Parse Evercode across 9 core facilities. Thread 🧵 #scRNAseq #singlecell #bioinformatics

New benchmarking paper on ONT's dorado RNA modification models, and the headline number is uncomfortable: 50-100% false discovery rate on real biological samples, even after filtering. Here is what that means for your experiments. academic.oup.com/nar/article/... #Epitranscriptomics #Nanopore

Systematic benchmarking of dorado basecalling models for RNA modification detection with highly multiplexed nanopore sequencing

Abstract. Nanopore direct RNA sequencing holds promise for advancing our understanding of the epitranscriptome. Recently, Oxford Nanopore Technologies rele

academic.oup.com

High-throughput sequencing has lived in core facilities for a decade. VITARI might be the first serious attempt to move it to the bench. Element Bio just announced their new platform. Here's what matters for sc/spatial labs. #SingleCell #SpatialTranscriptomics #NGS #Genomics #Bioinformatics

Python is another essential programming language in Bioinformatics... Here are 10 free Python 🐍 and Bioinformatics 🧬related resources to help you learning and practicing Python A 🧵 #Python #pythonprogramming #Bioinformatics

Bladder tumors are often classified as one molecular subtype. A new spatial atlas shows that classification can break down at the level that matters most: where the cancer actually is. aacrjournals.org/cancerdiscov... #BladderCancer #Oncology

Adding experimental data made RNA structure predictions worse. Not because the chemistry was wrong. Because the signal alignment upstream was noisy. That is the problem segSHAPE was built to fix. www.biorxiv.org/content/10.6... #Trinobia #ComputationalBiology

segSHAPE: RNA secondary structure prediction from nanopore direct RNA sequencing

RNAs adopt complex structures that regulate key biological processes, making accurate structure prediction essential. Chemical probing coupled with Nanopore direct RNA sequencing (DRS) offers a route ...

biorxiv.org

A pan-cancer single-cell atlas of tumor-associated dendritic cells just published in Nature Communications. Built by VIB and VUB. 178,000+ cells. 14 mouse models. 10 human cancer types. 🧬 www.nature.com/articles/s41... #Trinobia #Bioinformatics #scRNAseq

If you are running DESeq2 on APA-seq data, you are using the wrong tool for the job. A new Briefings in Bioinformatics paper builds the right one. doi.org/10.1093/bib/... #Bioinformatics #RNAseq #Transcriptomics

APAdeg enhances differentially expressed gene inference by leveraging site-specific signals in APA-seq data

Abstract. Alternative polyadenylation (APA) is a key post-transcriptional regulatory mechanism implicated in various diseases. Existing APA analysis tools

doi.org

1.1 million gut cells. 343 individuals. One of the largest single-cell atlases ever built for Crohn's disease. IBDverse is out in Nature Genetics, and the biology it surfaces is worth paying attention to. www.nature.com/articles/s41... #SingleCell #Bioinformatics #IBD

Single-cell RNA sequencing of terminal ileal biopsies identifies signatures of Crohn’s disease pathogenesis - Nature Genetics

IBDverse is a single-cell RNA atlas of terminal ileal biopsies comprising 1.1 million cells from 111 patients with Crohn’s disease and 232 healthy participants. Analysis of the data identifies genes, ...

nature.com

CRISPR screens tell you what genes matter. Spatial transcriptomics tells you where. Perturb-DBiT does both at once, in intact tissue, with 80,000+ guide libraries. Nature Biotechnology, June 2026. www.nature.com/articles/s41... #SpatialTranscriptomics #CRISPR #Bioinformatics

Large-scale, spatially resolved panoramic CRISPR screening in native tissue environments using Perturb-DBiT - Nature Biotechnology

In vivo CRISPR genetic perturbations are spatially mapped at scale.

nature.com

scRNA-seq loses cell morphology during dissociation. STAMP keeps it. A new mini review from St. Jude walks through the full downstream Python pipeline for STAMP data, linking gene expression directly to cell shape, size, and marker localization. 🧵 #SpatialTranscriptomics #Bioinformatics

You've run your scRNA-seq experiment. Now you want to know where your cells are heading. You reach for RNA velocity — but which method do you use, and does it actually matter? Turns out: yes, quite a lot. #RNAvelocity #scRNAseq #Bioinformatics #SingleCell #Genomics

Single LLMs can annotate cell types. They can also hallucinate them. 56% of initial predictions in a new study were classified as hallucinations. What if you make multiple LLMs deliberate until they agree? www.nature.com/articles/s42... #scRNAseq #Bioinformatics #LLM #AI

We just wrapped up a hands-on scRNA-seq data analysis workshop for PhD students at @dkfz.bsky.social 🎉 From experimental design to biological interpretation, researchers worked through the complete single-cell analysis journey. #scRNAseq #SingleCellRNAseq #Bioinformatics #ComputationalBiology

Most sequence-to-function models are trained on bulk tissue data. Decima changes that, trained on 22M+ single cells, it predicts gene expression from DNA sequence at cell type resolution. #Bioinformatics #scRNAseq #Genomics #MachineLearning #GeneRegulation

Learning R is an essential step for practicing bioinformatics. Here are 10 free resources to get you started 👇 #Rstats #Bioinformatics #Biology #Programming #Trinobia

1/7 Fungi kill ~1.7M people/year. They threaten enough crops to feed 4B people. Yet fungal infections remain one of the most underfunded areas in all of infectious disease biology. scRNA-seq is starting to change that. 🧵 #scRNAseq #FungalInfections

1/
Still a great intro to genomic surveillance:
 cambiotraining.github.io/sars-cov-2-g... #Trinobia #Bioinformatics #Genomics #NGS #GenomicSurveillance #Phylogenetics #Nextstrain #VariantCalling #ComputationalBiology #LifeSciences #Biotech #Research #Science #DataScience #MolecularBiology

Your RNA-seq differential expression results might be cleaner than you think. Or messier. The problem is you probably can't tell from the analysis alone. 🧵 Paper: academic.oup.com/bib/article/... #Bioinformatics #RNAseq #BatchEffects #DifferentialExpression #DeepLearning #ComputationalBiology

USADAE: a deep learning approach to disentangle hidden covariates in RNA-seq data

Abstract. Integrative analysis of RNA-seq datasets faces critical challenges in disentangling biologically meaningful signals from implicit confounders, su

academic.oup.com

Single-cell clustering sounds straightforward until you actually try it. Dropout, technical noise, and sparsity mean clean cluster boundaries are often artifacts of how you looked at the data. Paper: academic.oup.com/bib/article/27/3/bbaf169/8698827 #scRNAseq #Bioinformatics #CellClustering

scMVAF: a multi-view adaptive fusion clustering approach for single-cell RNA-sequencing data

Abstract. Single-cell RNA-sequencing (scRNA-seq) can excavate cellular heterogeneity and distinguish different types of cells. Clustering cells into subpop

academic.oup.com

7 web tools to explore genomics data you should know 🧵 #Bioinformatics #Genomics

Most RNA splicing models tell you what a sequence does. A new model from UPenn tells you what sequence you need to get the splicing outcome you want. elifesciences.org/articles/106... #RNAsplicing #GenomicsAI #Trinobia

Generative modeling for RNA splicing prediction and design

TrASPr+BOS enables accurate prediction and design of tissue-specific RNA splicing, even for tissues not trained on, uncovering unseen regulatory elements and guiding sequence edits that reshape splici...

elifesciences.org

Your RNA-seq results look clean. The statistics check out. The volcano plots look great. And you might still be completely wrong about the biology. Here is why that happens, and what to do about it. #SpatialTranscriptomics #SingleCellRNAseq #PancreaticCancer #Bioinformatics