Threadfin

Relationships between B-cell clones from paired single-cell RNA and receptor sequencing

Threadfin takes paired single-cell RNA and receptor (BCR or TCR) sequencing and describes every clone by the distribution of its cells' transcriptional states, so that clones, not cells, can be compared, mapped and tested.

Get started

1. Install

pip install threadfin                  # Python 3.10–3.14 on Linux, macOS or Windows; no GPU or R needed
pip install "threadfin[dandelion]"     # also installs Dandelion (0.5.x), to call clones from BCR or TCR contigs

Scanpy, Leiden, UMAP and Harmony are installed with it. Dandelion's clone calls depend on Python's hash seed: start Python with PYTHONHASHSEED=0 when clone ids must be identical run after run. Threadfin is validated with Dandelion 0.5.x; Dandelion 1.0 (October 2026) calls clones differently, so the extra installs sc-dandelion<1. The development version installs with pip install "git+https://github.com/Reynold4k/Threadfin.git".

2. Run

You need, for the same cells: raw gene counts in an AnnData object, the contig-level receptor table (Cell Ranger's filtered_contig_annotations.csv or an AIRR .tsv), and the donor and sample of each cell.

import scanpy as sc
import threadfin as tf

adata = sc.read_h5ad("b_cells.h5ad")           # raw counts, one row per cell

result = tf.run(
    adata,
    bcr="filtered_contig_annotations.csv",     # Cell Ranger VDJ output or an AIRR .tsv
    donor_key="donor",                         # the individual each cell came from
    sample_key="sample",                       # the sample or library each cell came from (never a sort gate,
                                               # tissue or time point you want to compare: those go in test=)
    n_jobs=8,                                  # CPU cores to use; the result is the same with any number
)

tf.run calls clones within each donor (Dandelion, paired chains), builds an expression embedding without the receptor genes, estimates one profile per clone, tests clonal coherence, lays out the clone map and looks for clonal programmes. On the 16 public datasets tested (up to 206,000 cells) it took 1 to 15 minutes with 8 cores.

3. Read the results

print(result.summary())                 # plain-language report
result.plot("threadfin_overview.pdf")   # clone map, coherence, stability, composition
print(result.methods())                 # a Methods paragraph with the settings and versions used
result.clones.to_csv("clones.csv")      # one row per clone: size, reliability, programme, map position

Common variations

• Your own labels (antigen binding, isotype, sort gate): any adata.obs columns, e.g. tf.run(..., test=["antigen_binding", "isotype"]).
• Time points or tissues, to ask whether clones keep their state: tf.run(..., time_key="timepoint").
• T cells: pass the TCR contigs as bcr; clones are called on identical beta and alpha junctions.
• scirpy users: pass your scirpy AnnData or MuData as bcr.
• Clones defined elsewhere: put them in adata.obs["clone_id"] and leave out bcr.
• Large data: n_jobs=-1 uses every CPU core and gives the same result; a million cells take about an hour on 8 cores.
• No data at hand: adata = tf.sim.simulate_repertoire() makes a repertoire with a known answer; it already has clone ids and an embedding, so run it with tf.run(adata, donor_key="donor", sample_key="context", basis="X_pca") (tutorial 1).

Tutorials

4.5.1 pins Dandelion below 1.0 and fixes T-cell contig files passed as a path. New in 4.5.0: T cells, scirpy input, result.methods() and gene-level summaries of the clone landscape (changelog).

What Threadfin does

Threadfin is a Python package for paired single-cell RNA and B-cell receptor (BCR) sequencing. It uses the clone — the B cells that descend from one V(D)J rearrangement — as the unit of analysis. Each clone is represented by the distribution of transcriptional states among its sampled cells, estimated relative to the cells sampled alongside it and shrunk according to the number of cells captured. Clones are then compared with one another: in a clone embedding with one point per clone, in clonal programmes of clones that behave alike, and in clone-level tests of anything you measured. Threadfin works with AnnData and Scanpy, and defines clones with Dandelion.

Why clones?

The cells of one clone are distributed across states: germinal-centre (GC) B cells, plasmablasts, plasma cells and memory B cells. Two clones can have the same average state and still differ in how their cells are distributed. Selection, antigen affinity and fate bias act on clones, so questions such as which clones connect the germinal centre with plasma-cell output or does antigen binding go with a clone's state need the clone as the unit of representation and of statistical inference. Three features of repertoires sampled in vivo make this difficult, and Threadfin addresses each:

• Most clones are captured as one or a few cells. Profiles are shrunk according to the number of cells sampled, and each clone carries a reliability score. No clone is discarded for being small.
• Clone membership is confounded with donor, sample, tissue and sorting gate. Each clone is described relative to the cells sampled with it, and every test shuffles labels within the sampling design.
• Treating cells as independent inflates significance. Tests use clones, not cells, as replicates.

What Threadfin answers

Examples from public data

The manuscript applies Threadfin to nine paired datasets from immunization, infection, vaccination and spontaneous germinal centres (figures); scripts and results for these and further public datasets are in case studies, and everything behind the manuscript — status, datasets, every analysis and a benchmark against related tools — is in manuscript/.

• Gut-associated lymphoid tissues (Siniscalco et al., Immunity 2025): the clone embedding resolves a continuous path of clones from GC B cells, through clones that contain both GC B cells and plasma cells, to plasma-cell clones, with clones that lack GC-fate labelling grouped at its end.
• Spontaneous germinal centres (Castrillon et al., eLife 2023): position along the clone embedding orders clones by the diversity of their cell states, and memory-containing clones concentrate at one end.
• Plasmodium infection (Skinner et al., Nat. Immunol. 2026): whether plasmablasts were sampled together with GC B cells of the same clone is a property of clones that cannot be read from the cell-level embedding.
• Model-antigen germinal centres (Merkenschlager et al., Nature 2025): the main axis of variation between clone profiles follows recent division in two cohorts whose sequencing libraries are crossed, so the association cannot be a library artefact; antigen-bait and dark-zone gates are also associated with clone profiles but were each sequenced in their own library.

A clone embedding that holds together

UMAP draws continuous populations of clones as one picture only for suitable settings, and the suitable settings differ between datasets. Threadfin therefore chooses them for each dataset by default: it lays the clones out over a ladder of settings, scores each layout for coherence (the fraction of clones in one connected piece) and for fidelity to the clone profiles, and keeps the most faithful setting that holds together in every layout it tried. A map is not allowed to be more fragmented than the clone profiles themselves, and the final map is checked once more before it is returned. The record of the choice is stored with the results, and the settings can always be fixed by hand (Tutorial 3).

tf.tl.embed_clones(adata)                               # automatic (default)
tf.tl.embed_clones(adata, n_neighbors=30, min_dist=0.3) # by hand
tf.tl.map_quality(features, xy)                         # score any clone map (profiles, coordinates)

How it works

1. Clones are defined from paired heavy- and light-chain sequences (for T cells, beta and alpha chains) within each donor (Dandelion).
2. Clone profiles. Each cell is described relative to the mean of its sampling context. The cells of a clone are summarised by a kernel mean embedding of their state distribution, so a clone divided between two states differs from a clone in an intermediate state, and the profile is shrunk towards the context according to the number of cells (best linear unbiased prediction under a random-effects model).
3. Clonal coherence is the intraclass correlation of state within clones, tested against clone labels permuted within samples.
4. The clone embedding is a UMAP of the clone profiles with settings chosen as above.
5. Clonal programmes are graph communities of clones; each split is tested against a single Gaussian group, and each programme's stability is estimated by bootstrap resampling.
6. Clone-level tests quantify how much of the variance between clone profiles a label explains, with labels permuted among the clones of each donor.

The full description, with references, is in Methods.

Documentation

Citation

A manuscript describing Threadfin is in preparation. Until it is available, please cite this repository (CITATION.cff) and the tools Threadfin builds on: Dandelion (Suo et al., Nat. Biotechnol. 2024) for clone definition, Scanpy (Wolf et al., Genome Biol. 2018) and UMAP (McInnes et al., J. Open Source Softw. 2018).

How Threadfin compares with related tools

Clonotrace, clone2vec, Benisse and CoNGA also combine lineage and transcriptome, and answer related but different questions. On the same data and clones (benchmark, with a one-page overview): Threadfin and Benisse represent every clone with two or more cells by default (clone2vec starts at three cells, Clonotrace at ten), and Threadfin's test of clonal coherence keeps its error rate when samples differ, where per-clone graph scores and cell-level tests do not. For recovering measured clone annotations Threadfin is comparable to clone2vec and to a simple mean of the cells' coordinates, and better than Clonotrace at its default and than Benisse; in simulation, clone2vec groups clones by their planted programme and tells mixed from in-between clones more accurately. Threadfin needs a fraction of the others' time (its clone profiles take seconds; clone2vec about 30 times longer, Clonotrace 300 to 2,400 times) and finished every run without help; a million cells take about an hour on 8 cores (scalability). The benchmark reports every result, including those where Threadfin is not the best.

Related tools

• Dandelion — single-cell immune receptor analysis; Threadfin's clone definition.
• Clonotrace — distances between expanded clones on the cell-state graph after label spreading.
• clone2vec — clone embeddings learned from which clones' cells are neighbours.
• Benisse — a joint space of BCR sequences and expression.
• CoNGA — graph-vs-graph analysis of T-cell receptor and expression similarity.
• Scirpy — single-cell immune receptor analysis in the scverse ecosystem; Threadfin reads its AnnData and MuData objects.
• Scanpy and AnnData — the data structures and single-cell tools Threadfin works with.
• Immcantation — B-cell receptor repertoire analysis, including clonal lineage reconstruction.

Support

Please report bugs and ask questions through GitHub issues.

License

MIT; see LICENSE.