Systematic benchmarking and optimal strategy selection of cross-species integration methods.
7 blocks of code here have a permanent identifier (oscr:b:…), listed with the matches below: citable for good, with a page of its own. Select a match and press Cite to copy it. What a block identifier is.
The 9 matches · 1 of them tie a paragraph to a whole file without given lines: a weak match, whose lines are not tinted
- [1] § Materials and methods › Dimension reduction and visualization ↔ core_method/ComBat.R, lines 75 to 143 · score 0.82 · RunPCA, FindClusters, FindNeighbors, DimPlot, RunUMAP, npcs · block
oscr:b:85R7TE6TZ7NPN-Q· Report a mismatch - [2] § Materials and methods › Dimension reduction and visualization ↔ core_method/R_for_all_tasks.R, lines 79 to 122 · score 0.82 · RunPCA, FindClusters, FindNeighbors, DimPlot, RunUMAP, npcs · block
oscr:b:AVM7V9T1RW0TC-4· Report a mismatch - [3] § Results › Construction and application of machine learning evaluation model ↔ model/Scripts/predict_overall_score.py, lines 2 to 61 · score 0.67 · Batch.ASW, kBET, Random Forest, predicting, training, NMI · block
oscr:b:V94XNPJ3F4S28-P· Report a mismatch - [4] § Materials and methods › Evaluated metrics › Local Inverse Simpson’s Index ↔ evaluation/ARI_LISI/evaluation_LISI_knn.py, lines 60 to 125 · score 0.60 · nearest neighbors, unique batches, Simpson, vectors, matrices, LISI · block
oscr:b:929CAE2J2WXTH-J· Report a mismatch - [5] § Materials and methods › Evaluated metrics ↔ model/Scripts/predict_overall_score.py, lines 2 to 61 · score 0.59 · graph connectivity, kBET, PCR, ARI, NMI, ASW · block
oscr:b:V94XNPJ3F4S28-P· Report a mismatch - [6] § Results › Multidimensional cross-species single-cell transcriptome integration framework ↔ evaluation/ARI_LISI/ARI_LISI_rds.R, lines 1 to 58 · score 0.56 · FastMNN, JointPCA, ComBat, Conos, RPCA, Harmony · block
oscr:b:1AHQ62NRNESWA-4· Report a mismatch - [7] § Results › Systematic assessment of factors affecting cross-species data integration ↔ evaluation/ARI_LISI/ARI_LISI_rds.R, lines 1 to 58 · score 0.54 · FastMNN, JointPCA, ComBat, RPCA, Harmony, CCA · block
oscr:b:1AHQ62NRNESWA-4· Report a mismatch - [8] § Materials and methods › Cross-species data integration methods › scVI and scANVI ↔ core_method/scANVI.py, the whole file · a weak match · score 0.53 · labels_key, latent, AnnData, scANVI, training, scVI · block
oscr:b:AVGH5F28KFD9G-O· Report a mismatch - [9] § Materials and methods › Evaluated metrics › NMI ↔ evaluation/ASW_NMI/evaluation_ASW_NMI.py, lines 138 to 211 · score 0.53 · Louvain clustering, scIB, NMI · block
oscr:b:JJ16WNV18QT5S-Q· Report a mismatch
Paper
Loaded from Europe PMC by your browser; OSCR does not store it: doi.org · Europe PMC
The paper is loaded when this pane is shown.
The authors' code
Python · 122 lines · 2.4 KB · CC-BY-4.0 · 2 matches
- import pandas as pd
- import numpy as np
- import joblib
- import argparse
- import sys
- parser = argparse.ArgumentParser(
- description="Predict Overall.Score for new integration methods using a trained Random Forest model"
- )
- parser.add_argument(
- "--model",
- required=True,
- help="Path to trained RF model (.joblib)"
- )
- parser.add_argument(
- "--input",
- required=True,
- help="CSV file with scIB metrics for new methods"
- )
- parser.add_argument(
- "--output",
- default="predicted_overall_score.csv",
- help="Output CSV file (default: predicted_overall_score.csv)"
- )
- parser.add_argument(
- "--train-data",
- required=True,
- help="Training data CSV used to fit the RF model (for OOD check)"
- )
- args = parser.parse_args()
- FEATURES = [
- "Graph.iLISI",
- "Batch.NMI",
- "Batch.ARI",
- "Batch.ASW",
- "PCR.batch",
- "kBET",
- "Graph.connectivity",
- "Graph.cLISI",
- "Cell.type.NMI",
- "Cell.type.ARI",
- "Cell.type.ASW",
- ]
- model = joblib.load(args.model)
- scaler = model.named_steps["scaler"]
- rf = model.named_steps["model"]
- new_data = pd.read_csv(args.input)
- missing = [f for f in FEATURES if f not in new_data.columns]
- if missing:
- print("Missing required features:")
- for f in missing:
- print(" -", f)
- sys.exit(1)
- X_new = new_data[FEATURES]
- new_data["Predicted.Overall.Score"] = model.predict(X_new)
- X_scaled = scaler.transform(X_new)
- tree_preds = np.array(
- [tree.predict(X_scaled) for tree in rf.estimators_]
- )
- new_data["Prediction.Std"] = tree_preds.std(axis=0)
- new_data["Prediction.Lower"] = (
- new_data["Predicted.Overall.Score"] - 2 * new_data["Prediction.Std"]
- )
- new_data["Prediction.Upper"] = (
- new_data["Predicted.Overall.Score"] + 2 * new_data["Prediction.Std"]
- )
- train_data = pd.read_csv(args.train_data)
- train_stats = train_data[FEATURES].describe().T[
- ["min", "max", "mean", "std"]
- ]
- ood_flags = []
- for i, row in X_new.iterrows():
- out_of_range = []
- for f in FEATURES:
- v = row[f]
- lo, hi = train_stats.loc[f, ["min", "max"]]
- if v < lo or v > hi:
- out_of_range.append(f)
- ood_flags.append(",".join(out_of_range))
- new_data["OOD.Features"] = ood_flags
- new_data["Is.OOD"] = new_data["OOD.Features"] != ""
- new_data.to_csv(args.output, index=False)
- cols_preview = [
- "Predicted.Overall.Score",
- "Prediction.Std",
- "Prediction.Lower",
- "Prediction.Upper",
- "Is.OOD",
- ]
- print(new_data[cols_preview].head())
predict_overall_score.py, under CC-BY-4.0 · at the source
Overview
- State Key Laboratory of Genetic Evolution and Animal Models, Kunming Institute of Zoology, Chinese Academy of Sciences, 17 Longxin Road, Kunming, Yunnan, 650201, China
- Kunming College of Life Science, University of Chinese Academy of Sciences, 19 Qingsong Road, Kunming, Yunnan, 650201, China
- Bio-X Center for Interdisciplinary Innovation, Yunnan University, 2 Cuihu North Road, Kunming, Yunnan 650091, China
Abstract
Single-cell RNA sequencing provides an unprecedented resolution for cellular heterogeneity and gene regulation, fostering cross-species comparative analyses with increasing interspecies data. However, integrating single-cell transcriptomic data faces challenges, including gene selection, evolutionary distance, and batch effects, with varying method performances. We utilized single-cell transcriptomic data from hippocampal tissues of seven mammals (e.g. mouse, human), evaluating 13 mainstream integration methods across 27 tasks with 11 metrics. To compare the performance of different methods, we developed a machine learning-based scoring model that assesses the contribution of each metric in a data-driven manner, thereby addressing the oversimplified assumptions of traditional manual weighting approaches. Our findings show that selecting highly variable one-to-one orthologous genes best balances species differences and commonalities. Most methods integrated closely related species, whereas scVI, a probabilistic model with distributions specified by deep neural networks, and its semi‑supervised extension scANVI, as well as the Seurat v5 method, which uses reciprocal principal component analysis (RPCAv5), effectively mapped distantly related species. Increased species numbers reduce gene overlap and heighten heterogeneity, increasing integration difficulty. The scANVI best maintained quality by balancing the biological signals and batch effect removal. Furthermore, we established an evaluation website to guide researchers in selecting the optimal integration methods for cross-species single-cell transcriptomic data analysis. Collectively, our findings provide a systematic, evidence-based framework that can assist researchers in rapidly selecting appropriate integration methods for cross-species single-cell transcriptomic studies.
Reproduced under the paper's license (CC BY), from the paper cited above.
Repository
Its files are read in the Code ↔ Paper reader above, with 9 matches between paragraphs and lines of code.
figshare 31976259
Availability: 1 check, the latest on 26 September 2026: the link answers (HTTP 200)
- 26 September 2026: the link answers (HTTP 200)
37 files
- core_method/
BBKNN.py , Python, 117 lines - core_method/
CCA.R , R, 75 lines - core_method/
ComBat.R , R, 143 lines, 1 match - core_method/
Conos.R , R, 126 lines - core_method/
FastMNN.R , R, 76 lines - core_method/
Harmony.R , R, 75 lines - core_method/
JointPCA.R , R, 73 lines - core_method/
RPCA.R , R, 78 lines - core_method/
R_for_all_tasks.R , R, 123 lines, 1 match - core_method/
Scanorama.py , Python, 117 lines - core_method/
python_for_all_tasks.py , Python, 67 lines - core_method/
scANVI.py , Python, 45 lines, 1 match - core_method/
scVI.py , Python, 57 lines - evaluation/
ARI_LISI/ , R, 39 linesARI_LISIH5.R - evaluation/
ARI_LISI/ , Python, 48 linesARI_LISIH5pre.py - evaluation/
ARI_LISI/ , R, 122 lines, 2 matchesARI_LISI_rds.R - evaluation/
ARI_LISI/ , R, 277 linesevaluation_ARI_LISI_R_F. R - evaluation/
ARI_LISI/ , R, 226 linesevaluation_ARI_LISI_py_F .R - evaluation/
ARI_LISI/ , Python, 193 lines, 1 matchevaluation_LISI_knn.py - evaluation/
ARI_LISI/ , R, 96 linessamap_ARI.R - evaluation/
ARI_LISI/ , Python, 57 linessamap_ARIpre.py - evaluation/
ARI_LISI/ , Python, 37 linessamap_LISI.py - evaluation/
ARI_LISI/ , Python, 37 linessamap_LISIpre.py - evaluation/
ASW_NMI/ , Python, 61 linesASW_NMI02.py - evaluation/
ASW_NMI/ , Python, 294 lines, 1 matchevaluation_ASW_NMI.py - evaluation/
ASW_NMI/ , Python, 135 linessamap_ASW_NMI.py - evaluation/
ASW_NMI/ , Python, 38 linessamap_NMIpre.py - evaluation/
scib/ , Python, 120 linesH5_evaluation.py - evaluation/
scib/ , R, 73 linesexport_all_embeddings.R - evaluation/
scib/ , Python, 66 linessamap_scibmetric.py - evaluation/
scib/ , Python, 125 linessaturn_evaluation.py - evaluation/
scib/ , Python, 166 linesscib_ccevaluation.py - evaluation/
scib/ , Python, 146 linesscib_evaluation.py - model/
Scripts/ , Python, 54 linesknn.py - model/
Scripts/ , Python, 122 lines, 2 matchespredict_overall_score.py - model/
Scripts/ , Python, 146 linestest_prediction.py - README.md, Text, 83 lines
The paper's code and data availability statement is in the Data section.
Tracing map
Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.
What the map holds:
- 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
- 36 scripts, each with its path and the digest of its content;
- 9 matches between paragraphs of the paper and lines of the code (method lexical-v1);
- neither the text of the paper nor the code itself.
Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.
Data
Datasets cited
- figshare:31968882, at figshare; found in “Data availability”
Data availability
Datasets were collected from published studies as described in Supplementary Tables S1 and S2. The preprocessed datasets for each task are available at https://
Reproduced under the paper's license (CC BY), from the paper cited above.
Versions
The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.
Version 1, 27 September 2026: the first record
Recorded: type, language, journal, volume, issue, pages, dates, 6 authors, 3 keywords, 13 MeSH terms, 3 funders, 65 references.
Cite
This paper
Wang, R., Zheng, J., Mao, C., Zhang, Y.-P., Ding, Z., & Wang, G.-D. (2026). Systematic benchmarking and optimal strategy selection of cross-species integration methods. Briefings in bioinformatics, 27(5), bbag490. https://
BibTeX
@article{wang2026systema
author = {Wang, Ruolin and Zheng, Junjuan and Mao, Chuning and Zhang, Ya-Ping and Ding, Zhaoli and Wang, Guo-Dong},
title = {{Systematic benchmarking and optimal strategy selection of cross-species integration methods}},
journal = {Briefings in bioinformatics},
year = {2026},
month = sep,
volume = {27},
number = {5},
pages = {bbag490},
publisher = {Oxford University Press},
issn = {1467-5463},
doi = {10.1093/
url = {https://
pmid = {42721443},
pmcid = {PMC13561306}
}
RIS
TY - JOUR
AU - Wang, Ruolin
AU - Zheng, Junjuan
AU - Mao, Chuning
AU - Zhang, Ya-Ping
AU - Ding, Zhaoli
AU - Wang, Guo-Dong
TI - Systematic benchmarking and optimal strategy selection of cross-species integration methods
T2 - Briefings in bioinformatics
J2 - Brief Bioinform
PY - 2026
DA - 2026/
VL - 27
IS - 5
SP - bbag490
SN - 1467-5463
PB - Oxford University Press
DO - 10.1093/
UR - https://
LA - en
ER -
CSL-JSON
{
"id": "10.1093/
"type": "article-journal",
"title": "Systematic benchmarking and optimal strategy selection of cross-species integration methods",
"container-title": "Briefings in bioinformatics",
"author": [
{
"family": "Wang",
"given": "Ruolin"
},
{
"family": "Zheng",
"given": "Junjuan"
},
{
"family": "Mao",
"given": "Chuning"
},
{
"family": "Zhang",
"given": "Ya-Ping"
},
{
"family": "Ding",
"given": "Zhaoli"
},
{
"family": "Wang",
"given": "Guo-Dong"
}
],
"container-title-short":
"volume": "27",
"issue": "5",
"page": "bbag490",
"DOI": "10.1093/
"PMID": "42721443",
"PMCID": "PMC13561306",
"ISSN": "1467-5463",
"publisher": "Oxford University Press",
"URL": "https://
"language": "en",
"issued": {
"date-parts": [
[
2026,
9,
1
]
]
}
}
The tracing map gets a citation of its own once an author has validated it and it has a DOI.
Similar papers
The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.
- [1] doi:10.1038/s41591-024-03150-z [code]
- A brain cell atlas integrating single-cell transcriptomes across human brain regions.Journal: Nature medicineIn common: anndata, Scanpy, igraph, 11 other tools, methods / tools, genetics / omics, mouse, 9 references
- [2] doi:10.1038/s42255-022-00657-y [code]
- HypoMap-a unified single-cell gene expression atlas of the murine hypothalamus.Journal: Nature metabolismIn common: anndata, Scanpy, igraph, 9 other tools, mouse, 8 references
- [3] doi:10.1038/s41588-024-01961-x [code]
- Spatial and single-nucleus transcriptomic analysis of genetic and sporadic forms of Alzheimer's disease.Journal: Nature geneticsIn common: anndata, Scanpy, igraph, 10 other tools, genetics / omics, mouse, 6 references
- [4] doi:10.1016/j.xgen.2022.100107 [code]
- Single nucleus multi-omics identifies human cortical cell regulatory genome diversity.Journal: Cell genomicsIn common: anndata, Scanpy, igraph, 11 other tools, genetics / omics, 6 references
- [5] doi:10.1016/j.xgen.2024.100739 [code]
- CSI-GEP: A GPU-based unsupervised machine learning approach for recovering gene expression programs in atlas-scale single-cell RNA-seq data.Journal: Cell genomicsIn common: anndata, Scanpy, igraph, 11 other tools, genetics / omics, mouse, 3 references
- [6] doi:10.1016/j.crmeth.2023.100498 [code]
- hdWGCNA identifies co-expression networks in high-dimensional transcriptomics data.Journal: Cell reports methodsIn common: anndata, Scanpy, igraph, 10 other tools, genetics / omics, 6 references
- [7] doi:10.1016/j.cell.2025.06.031 [code]
- Single-cell multiregion epigenomic rewiring in Alzheimer's disease progression and cognitive resilience.Journal: CellIn common: anndata, Scanpy, igraph, 11 other tools, genetics / omics, 4 references
- [8] doi:10.1038/s41593-025-02022-0 [code]
- An integrated single-nucleus and spatial transcriptomics atlas reveals the molecular landscape of the human hippocampus.Journal: Nature neuroscienceIn common: anndata, Scanpy, igraph, 10 other tools, genetics / omics, 5 references
- [9] doi:10.1038/s41586-023-06824-9 [code]
- Single-cell analysis of chromatin accessibility in the adult mouse brain.Journal: NatureIn common: anndata, Scanpy, igraph, 10 other tools, genetics / omics, mouse, 4 references
- [10] doi:10.1038/s41467-021-25534-2 [code]
- Learning interpretable cellular and gene signature embeddings from single-cell transcriptomic data.Journal: Nature communicationsIn common: anndata, Scanpy, Seurat, 7 other tools, genetics / omics, mouse, 8 references
Contribute
The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.
Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.
Claim this paper
Correct its record
Say what each link of this record is, remove the ones that are not the paper's, add the ones that are missing. The correction becomes a new version of the record, in its Versions section.
Validate its tracing map
You validate the map as this page shows it: 1 repository of the authors' code, each at its verified commit and with its license, 36 scripts, and 9 matches between paragraphs and code (see the Code and Map sections). It then receives a DOI on Zenodo, with you (your ORCID iD) and OSCR as its creators; the code itself is not deposited.
The map's fingerprint: sha256:a8bd5a9b65eab76c…
Add the badge to the code's README
If you are an author or a maintainer of this paper's code, you can add this badge. It says what OSCR knows of the paper (its code, its validated tracing map) and links the code to this page. Paste one of these snippets under the README's title. Keep the two marks around it, so that the badge can be replaced in place and is never added twice. OSCR changes nothing in your repository.
Markdown
<!-- oscr-badge:start -->
[](https://openscicode.org/badge/doi/10.1093/bib/bbag490)
<!-- oscr-badge:end -->reStructuredText
.. oscr-badge:start
.. image:: https://openscicode.org/badge/doi/10.1093/bib/bbag490.svg
:target: https://openscicode.org/badge/doi/10.1093/bib/bbag490
:alt: OSCR
.. oscr-badge:endHTML
<!-- oscr-badge:start -->
<a href="https://openscicode.org/badge/doi/10.1093/bib/bbag490"><img src="https://openscicode.org/badge/doi/10.1093/bib/bbag490.svg" alt="OSCR"></a>
<!-- oscr-badge:end -->If you choose to, you can have the badge kept up to date for you, with oscr badge --write from the command line or with the GitHub Action in your repository's workflows. Every other form and option (a card, AsciiDoc, LaTeX, shields.io) is on the badge page.
Request its removal
To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).
Discussion, reproductions, activity
Discussion: code errors and mismatches between the code and the paper about this paper and its code, from readers and its authors, beside the paper and its tracing map.
Report a mismatch between the code and the paper · Report a code error (signed in)
Reproductions: reports from readers who ran the authors' code and did not get the paper's result, what they ran, with which environment, commit and data.
Report a reproduction failure (signed in)
Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.
