Plate Nº 95 · recorded November 18, 2026
Biology & EvolutionReported finding
iCellOmics unifies multiomics data for 17 industrial cell lines
iCellOmics launches with 17 industrial cell lines, 170,408,888 DNA variants and 37,213 literature-derived gene–phenotype links, plus tools for cross-omics searches.
By Elena Vasquez4 min read885 words
In brief
- The current iCellOmics release covers 17 industrially relevant cell lines and six data categories.
- It holds 170,408,888 genomic variants from 55 whole-genome projects across 15 cell lines.
- Bulk transcriptomes include 673 differential-expression comparisons from 241 projects; single-cell data cover 1,047,274 cells from 30 projects.
- Chromatin-accessibility data include 287,089,415 peaks from 171 ATAC-seq projects.
- Researchers extracted 37,213 gene–phenotype relationships from 6,312 publications; the paper’s online date is Nov. 10 and its journal listing is 2026.

The current iCellOmics release brings together 17 industrially relevant cell lines and 170,408,888 genomic variants with functional annotations in one open-access multiomics database.
Multiomics means combining measurements from several biological layers instead of studying DNA alone. Industrial cell lines are grown for research or production. Some directly produce recombinant proteins, vaccines and viral vectors; others serve as biomedical models used in comparative studies.
Host-cell engineering means deliberately changing those cells to improve production or study their behavior. These lines support development and worldwide supply of recombinant therapeutics, viral vaccines and gene therapies.
What does the database contain?
The National Genomics Data Center (NGDC), part of the China National Center for Bioinformation (CNCB), released iCellOmics.
Xiaojing Gao and colleagues published the work in Nucleic Acids Research as iCellOmics: connecting multi-omics profiles and phenotype evidence across industrially relevant cell lines. The journal lists the paper for 2026, with an online publication date of Nov. 10. Its DOI is 10.1093/nar/gkag1148.
According to the authors, iCellOmics is the first resource to integrate multiomics data and literature-curated gene–phenotype knowledge for multiple industrially relevant cell lines within one analysis framework. The priority claim comes from the research team.
The current release combines six data categories:
- Reference genomes and annotations: The collection covers seven species and 144,018 protein-coding genes, which carry instructions for making proteins.
- Genomic variants: The database lists 170,408,888 DNA sequence differences from 55 whole-genome sequencing projects across 15 cell lines, along with functional annotations, or explanatory labels.
- Bulk transcriptomes: It provides 673 differential-expression comparisons from 241 projects measuring changes in gene activity across groups of cells.
- Single-cell transcriptomes: The collection covers 1,047,274 individual cells from 30 projects.
- Chromatin accessibility: The database includes 287,089,415 peaks from 171 ATAC-seq projects; each peak marks a DNA region accessible to cellular machinery.
- Gene–phenotype evidence: The collection contains 37,213 relationships from 6,312 publications, involving 6,605 gene or protein entities and 104 standardized phenotype subcategories.
A reference genome is a baseline DNA sequence used to identify differences, while annotations are labels attached to genes or variants.
Transcriptome data record which genes are active; single-cell data separate those signals by cell. Chromatin is the DNA-protein material that packages genes, and accessibility indicates which DNA regions are available for regulation. A whole-genome project reads the full DNA sequence.
In plain terms, iCellOmics links DNA differences, gene activity, regulatory access and published traits within the same 17 cell lines.
How can researchers use iCellOmics?
Researchers can browse data in a cell-line-centered view across all modules. The Gene Search function provides gene-centric cross-omics retrieval, meaning it assembles several data types for one gene in one cell line. The profile lists variant sites, expression comparisons, accessible chromatin regions, a single-cell transcriptome overview and gene–phenotype relationships from the literature.
The literature evidence also appears in an interactive knowledge graph, a network that links genes or proteins with observable traits. Users can switch between detailed and collapsed views, then filter by cell line, phenotype category, entity name, relationship type or PubMed ID, the identifier assigned to an article in the PubMed database.
Each relationship supplies relevant literature context, the experimental changes tested and their reported effects.
The database integrates the JBrowse 2 genome browser for inspecting genome locations and online Gene Ontology and KEGG enrichment tools, which test whether groups of genes cluster around known functions or pathways. Researchers can also download processed reference-genome, chromatin-accessibility and single-cell-transcriptome datasets for local analysis.
Together, these features offer knowledge-guided support for host-cell engineering. For users, the practical gain is a faster route from a cell line or gene to relevant evidence.
Why does integrating these records matter?
Before this release, genomic, transcriptomic and epigenomic profiles for industrial cell lines sat in separate repositories, projects and publications. Phenotype information extracted from scientific papers was scattered as well.
Different reference genomes and analytical workflows complicate comparisons because each dataset follows different assumptions. Transcriptomic profiles track gene activity; epigenomic profiles track DNA packaging and regulation. That fragmentation made systematic characterization difficult.
iCellOmics places those records into shared modules and connects them to evidence about cell traits. The authors say the resource can support basic research, host-cell engineering, and development and optimization of high-yield, stable producer cell lines.
The consolidation should make it easier to compare cell lines and trace biological evidence behind proposed engineering changes.
What should researchers check before relying on iCellOmics?
The release is broad, but it remains a map of existing evidence rather than a new laboratory experiment. Its phenotype layer depends on extracting relationships from 6,312 publications, while its other layers inherit references, sampling and processing from contributing sequencing projects.
The announcement does not quantify curation error, independent accuracy or how evenly coverage is distributed among the 17 lines.
Coverage is also bounded: this release covers 17 lines and six data categories. The underlying studies still matter. Researchers should inspect source studies, apply their own quality checks and verify that a trait or pathway holds in their chosen line before making operational decisions.
The practical takeaway is simple: use iCellOmics to find and compare evidence, not to replace laboratory validation.
via phys.org/rss-feed/ (Source)
More from Elena Vasquez
Show full bio
Correspondent covering money, funding and market shifts around general science news at SciBeat.
503 articles
Nearby plates
- Five chronic illnesses may share a common biological root, study finds
- New Immune Cell Atlas Links Genetic Variants to Disease Mechanisms
- MIT team builds method that lets living cells mail out their own RNA
- Missing Gut Protein May Prime Children for Peanut Allergy
- Google DeepMind Embeds Invisible Watermarks in AI-Designed Proteins