CytokineFindeR: an R-package for benchmarking methods and databases for identifying cytokines
Tang, J. S.; Singh, A.; Singh, A.
Show abstract
Cytokines play a central role in disease but are hard to study due to their short half-life and low abundance. Computational approaches have utilized gene-expression for pinpointing cytokine drivers of diseases. We benchmark various cytokine identification methods and found little congruency between gene sets for the same ligand, and variable performance in identifying cytokines based on curated ligand-receptor interactions. CytoSig, a model-based approach, was generally better at identifying the correct disease cytokine but failed for certain cytokines such as IL-13. We developed CytokinefindeR, an R package that enables comparative analysis of cytokine detection across 10 databases and four computational approaches.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- RIDDEN: Data-driven inference of receptor activity from transcriptomic data 94%
- STREAK: A Supervised Cell Surface Receptor Abundance Estimation Strategy for Single Cell RNA-Sequencing Data using Feature Selection and Thresholded Gene Set Scoring 94%
- GeneCOCOA: Detecting context-specific functions of individual genes using co-expression data 93%
Similar papers in this journal
Similar papers in this journal
- Enhancing Gene Set Overrepresentation Analysis with Large Language Models 93%
- SPECK: An Unsupervised Learning Approach for Cell Surface Receptor Abundance Estimation for Single Cell RNA-Sequencing Data 92%
- Enhanced Annotation of CD45RA to Distinguish T cell Subsets in Single Cell RNA-seq via Machine Learning 91%
Similar papers in this journal
- DeltaNeTS+: Elucidating the mechanism of drugs and diseases using gene expression and transcriptional regulatory networks 93%
- GEOlimma: Differential Expression Analysis and Feature Selection Using Pre-Existing Microarray Data 92%
- Single sample pathway analysis in metabolomics: performance evaluation and application 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.