Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Mechanistic Models: Compartment Models in Individual and Population Analysis01:23

Mechanistic Models: Compartment Models in Individual and Population Analysis

Mechanistic models are utilized in individual analysis using single-source data, but imperfections arise due to data collection errors, preventing perfect prediction of observed data. The mathematical equation involves known values (Xi), observed concentrations (Ci), measurement errors (εi), model parameters (ϕj), and the related function (ƒi) for i number of values. Different least-squares metrics quantify differences between predicted and observed values. The ordinary least squares (OLS)...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A Cross-Sectional Study on Baseline Characteristics and Exposure Spectrum of Occupational Lead and Coexisting Heavy Metal-Exposed Populations.

Journal of applied toxicology : JAT·2026
Same author

Precise programmable tumor cell subpopulation sorting <i>via</i> an electromagnetic microfluidic platform.

Lab on a chip·2026
Same author

Effects of biochar derived from different feedstocks on soil microbial nutrient limitation in a <i>Phyllostachys edulis</i> forest.

Ying yong sheng tai xue bao = The journal of applied ecology·2026
Same author

Altered heart rate variability is associated with all-cause mortality in patients with co-morbid insomnia and obstructive sleep apnea.

Sleep·2026
Same author

Dual mode analysis of lead ions in regulation of intestinal flora <i>via</i> split-DNA re-assembly catalyzed colorimetric reaction and <i>trans</i>-cleavage activity of Cas12a/crRNA.

Analytical methods : advancing methods and applications·2026
Same author

Clinic-based characteristics of animal-related injury presentations in Hangzhou, China: a 10-year retrospective analysis.

Frontiers in public health·2026

Related Experiment Video

Updated: Jun 14, 2026

A Versatile Automated Platform for Micro-scale Cell Stimulation Experiments
12:21

A Versatile Automated Platform for Micro-scale Cell Stimulation Experiments

Published on: August 6, 2013

10.7K

BioLLM: A standardized framework for integrating and benchmarking single-cell foundation models.

Ping Qiu1,2, Qianqian Chen2, Hua Qin2

  • 1College of Life Sciences, University of Chinese Academy of Sciences, Beijing 100049, China.

Patterns (New York, N.Y.)
|August 22, 2025
PubMed
Summary

BioLLM unifies single-cell foundation models (scFMs) for RNA sequencing analysis, simplifying access and benchmarking. Evaluation reveals scGPT excels across tasks, while Geneformer and scFoundation show gene-level strengths.

Keywords:
model benchmarkingsingle-cell foundation modelsunified frameworkzero shot and fine-tuning

More Related Videos

Workflow for High-content, Individual Cell Quantification of Fluorescent Markers from Universal Microscope Data, Supported by Open Source Software
09:57

Workflow for High-content, Individual Cell Quantification of Fluorescent Markers from Universal Microscope Data, Supported by Open Source Software

Published on: December 16, 2014

13.1K
A High-throughput Cell Microarray Platform for Correlative Analysis of Cell Differentiation and Traction Forces
12:04

A High-throughput Cell Microarray Platform for Correlative Analysis of Cell Differentiation and Traction Forces

Published on: March 1, 2017

9.8K

Related Experiment Videos

Last Updated: Jun 14, 2026

A Versatile Automated Platform for Micro-scale Cell Stimulation Experiments
12:21

A Versatile Automated Platform for Micro-scale Cell Stimulation Experiments

Published on: August 6, 2013

10.7K
Workflow for High-content, Individual Cell Quantification of Fluorescent Markers from Universal Microscope Data, Supported by Open Source Software
09:57

Workflow for High-content, Individual Cell Quantification of Fluorescent Markers from Universal Microscope Data, Supported by Open Source Software

Published on: December 16, 2014

13.1K
A High-throughput Cell Microarray Platform for Correlative Analysis of Cell Differentiation and Traction Forces
12:04

A High-throughput Cell Microarray Platform for Correlative Analysis of Cell Differentiation and Traction Forces

Published on: March 1, 2017

9.8K

Area of Science:

  • Computational Biology
  • Genomics
  • Bioinformatics

Background:

  • Single-cell foundation models (scFMs) offer powerful tools for analyzing single-cell RNA sequencing (scRNA-seq) data.
  • Heterogeneous architectures and coding standards pose significant challenges to the widespread application and evaluation of scFMs.
  • A unified framework is needed to streamline the integration and utilization of diverse scFMs in biological research.

Purpose of the Study:

  • To introduce BioLLM, a novel unified framework designed to integrate and standardize the application of scFMs for scRNA-seq analysis.
  • To provide a standardized interface and benchmarking system for evaluating the performance of various scFMs.
  • To assess the capabilities and limitations of prominent scFMs, including scGPT, Geneformer, scFoundation, and scBERT.

Main Methods:

  • Development of BioLLM, a framework offering a unified interface for accessing and applying multiple scFMs.
  • Implementation of standardized APIs and comprehensive documentation within BioLLM to ensure consistent model usage and switching.
  • Conducting a comprehensive evaluation of selected scFMs across various scRNA-seq analysis tasks, including zero-shot and fine-tuning scenarios.

Main Results:

  • BioLLM successfully integrates diverse scFMs, overcoming architectural and coding inconsistencies for streamlined analysis.
  • scGPT demonstrated robust performance across all evaluated tasks, highlighting its versatility in scRNA-seq analysis.
  • Geneformer and scFoundation showed strong performance in gene-level tasks, attributed to effective pretraining strategies, while scBERT exhibited limitations likely due to model size and data constraints.

Conclusions:

  • BioLLM provides a crucial tool for the scientific community, enabling easier access to and consistent evaluation of scFMs.
  • The framework facilitates a deeper understanding of scFM performance, guiding researchers in selecting appropriate models for their specific analyses.
  • BioLLM empowers researchers to fully leverage the potential of foundational models, advancing the study of complex biological systems through enhanced single-cell data interpretation.