Biology subjects

Hunewald, O.

Publications and source records attributed to Hunewald, O..

3 recordsLinked to original sources

bioRxiv2026

Language models annotate mass cytometry clusters within the range where human experts disagree

Cell-type identification of clusters from unsupervised cytometry pipelines (e.g. FlowSOM, GigaSOM) is a time-consuming, expert-dependent bottleneck, typically requiring an expert to read a heatmap of median marker intensities and assign a label to each cluster. Large language models (LLMs) can annotate single-cell RNA-seq clusters from marker-gene lists, but cytometry cluster data differ in both dimensionality and data type. We present a direct procedure in which cluster medians are scaled per marker and grouped by similarity into a few dozen metaclusters. The continuous marker profiles are then passed to the LLM, which returns a cell-type label and confidence level for each metacluster in a single call, without a gating hierarchy, manually defined marker-specific thresholds, or task-specific training data. Across four human and mouse datasets spanning blood, bone marrow and gut, a single well-chosen model reproduces the reference annotation at the major-lineage level with 85-98% metacluster-level concordance (cloud Claude Sonnet 4.6 85-98%, best local model 83-96%), substantially exceeding majority-class baselines; chance-corrected agreement is Cohen's k = 0.67-0.83. Local open-weight models perform competitively with the cloud model on the independent benchmarks, enabling application to clinical or patient-derived data subject to data-protection restrictions. Reliability is model-dependent: no local model is best everywhere, and the frontier model Claude Opus 5 was no more reliable than Claude Sonnet 4.6, the most consistent in concordance across datasets. Because no annotation is absolute ground truth, we obtained two additional expert annotations of the same continuous heatmaps, blinded to the reference and to all model output, to establish a method-matched agreement ceiling. The experts agreed on ~99% of major-lineage assignments but only ~73% of T-cell maturation states; on that axis the best models fall within the expert disagreement range. LLM-based annotation is therefore a practical, training-free alternative to rule-based annotation, provided its output is checked against the intended panel and species.

bioinformatics↗

bioRxiv2020

GigaSOM.jl: High-perfomance clustering and visualization of huge cytometry datasets

BackgroundThe amount of data generated in large clinical and phenotyping studies that use single-cell cytometry is constantly growing. Recent technological advances allow to easily generate data with hundreds of millions of single-cell data points with more than 40 parameters, originating from thousands of individual samples. The analysis of that amount of high-dimensional data becomes demanding in both hardware and software of high-performance computational resources. Current software tools often do not scale to the datasets of such size; users are thus forced to down-sample the data to bearable sizes, in turn losing accuracy and ability to detect many underlying complex phenomena. ResultsWe present GigaSOM.jl, a fast and scalable implementation of clustering and dimensionality-reduction for flow and mass cytometry data. The implementation of GigaSOM.jl in the high-level and high-performance programming language Julia makes it accessible to the scientific community, and allows for efficient handling and processing of datasets with billions of data points using distributed computing infrastructures. We describe the design of GigaSOM.jl, measure its performance and horizontal scaling capability, and showcase the functionality on a large dataset from a recent study. ConclusionsGigaSOM.jl facilitates utilization of the commonly available high-performance computing resources to process the largest available datasets within minutes, while producing results of the same quality as the current state-of-art software. Measurements indicate that the performance scales to much larger datasets. The example use on the data from an massive mouse phenotyping effort confirms the applicability of GigaSOM.jl to huge-scale studies. Key pointsO_LIGigaSOM.jl improves the applicability of FlowSOM-style single-cell cytometry data analysis by increasing the acceptable dataset size to billions of single cells. C_LIO_LISignificant speedup over current methods is achieved by distributed processing and utilization of efficient algorithms. C_LIO_LIGigaSOM.jl package includes support for fast visualization of multidimensional data. C_LI

bioinformatics↗

bioRxiv2017

High-throughput sequencing of murine immunoglobulin heavy chain transcripts using single side unique molecular identifiers on an Ion Torrent PGM

With the advent of high-throughput sequencing (HTS), profiling immunoglobulin (IG) repertoires has become an essential part of immunological research. Advances in sequencing technology enable the IonTorrent Personal Genome Machine (PGM) to cover the full-length of IG mRNA transcripts. Nucleotide insertions and deletions (indels) are the dominant errors of the PGM sequencing platform and can critically influence IG repertoire assessments. Here, we present a PGM-tailored IG repertoire sequencing approach combining error correction through unique molecular identifier (UID) barcoding and indel detection through ImMunoGeneTics (IMGT), the most commonly used sequence alignment database for IG sequences. Using artificially falsified sequences for benchmarking, we found that IMGT efficiently detects 98% of the introduced indels through gene-segment frameshifts. Undetected indels are either located at the ends of the sequences or produce masked frameshifts with an insertion and deletion in close proximity. IMGTs indel correction algorithm resolves up to 87% of the tested insertions, but no deletions. The complementary determining regions 3 (CDR3s) are returned 100% correct for up to 3 insertions or 3 deletions through conservative culling. We further show, that our PGM-tailored unique molecular identifiers results in highly accurate HTS datasets if combined with the presented data processing. In this regard, considering sequences with at least two copies from datasets with UID families of minimum 3 reads result in correct sequences with over 99% confidence. The protocol and sample processing strategies described in this study will help to establish benchtop-scale sequencing of IG heavy chain transcripts in the field of IG repertoire research.

immunology↗