Coexpression Network · maomlab
CryptoCEN
A co-expression network for Cryptococcus neoformans to predict gene function and prioritise drug targets.
What CryptoCEN is
Cryptococcus neoformans is a widespread human fungal pathogen that has substantially diverged from model yeasts. Approximately 60% of its predicted genes lack functional annotations, hindering research into its biology and the identification of therapeutic vulnerabilities.
CryptoCEN is a co-expression network generated from publicly available transcriptomic RNA-sequencing data. It operates on the principle of guilt-by-association—where genes demonstrating correlated expression patterns across diverse experimental conditions are likely involved in shared biological pathways—to uncover gene functions and reconstruct protein networks.
The dataset includes expression estimates, filtered and unfiltered co-expression associations, functional annotations from FungiDB, and orthology mappings to Saccharomyces cerevisiae protein complexes, as well as sequence similarity and co-evolution networks.
How it was made
Estimated gene expression profiles were computed across a large collection of publicly available RNA-seq runs for Cryptococcus neoformans strain H99. Pairwise correlation metrics were derived across these expression runs to construct the co-expression network.
The data release provides the full co-expression network alongside filtered subsets, such as the top 50 co-expressed partners per gene and interactions meeting a score threshold above 0.95. These are packaged alongside protein sequence similarity networks derived from BlastP, co-evolution networks, and protein feature annotations sourced from FungiDB Release 49.
Key facts
- Target organism
- Cryptococcus neoformans H99 (NCBI Taxon: 235443)
- Dataset size category
- 10M to 100M rows
- Primary file format
- TSV
- External annotation source
- FungiDB Release 49
- Network types included
- Co-expression, BlastP sequence similarity, and Co-evolution
What you can use it for
- Unannotated gene function prediction
- Infer biological pathways and functions for uncharacterised C. neoformans genes using guilt-by-association with known co-expressed partners.
- DNA damage response discovery
- Identify candidate genes involved in DNA damage repair mechanisms based on network associations with established repair factors.
- Antifungal drug target prioritisation
- Prioritise essential or pathway-critical fungal pathogen proteins for antimicrobial investigation and drug development.
- Multi-network comparative analysis
- Cross-reference co-expression links against BlastP sequence similarity and co-evolutionary patterns across fungal genomes.
Explore the data
The schema, real rows — read in your browser, nothing downloaded.
Explore
Read the real rows without downloading anything
Reading rows…
LiveRead from maomlab/CryptoCEN at the moment you asked. Nothing is cached or stored — every row above came from that request.
Strengths and limits
Good at
- Directly targets the 60% functional annotation deficit in Cryptococcus neoformans.
- Demonstrated 100% predictive accuracy when experimentally testing 13 newly identified DNA damage response genes.
- Includes pre-filtered files with defined association thresholds alongside the full-scale network for ease of querying.
- Integrates orthology data connecting C. neoformans genes to known Saccharomyces cerevisiae protein complexes.
Watch out for
- Gene models and feature annotations are tied specifically to C. neoformans strain H99 and FungiDB Release 49.
- Co-expression indicates shared regulation and pathway membership rather than direct physical binding or enzymatic activity.
Questions
Can I look inside CryptoCEN without downloading it?
Yes. This page reads its schema, real rows and per-column statistics in the browser. Nothing is downloaded to your machine.
What licence is CryptoCEN under?
Its authors declare mit. Read the terms on the original page before reusing the data.
Which Cryptococcus strain is used in this dataset?
The annotations and transcript models are based on Cryptococcus neoformans var. grubii H99 (NCBI Taxon: 235443).
What pre-filtered interaction sets are provided?
The resource contains a dataset listing the top 50 co-expression associations for each gene, as well as a more stringent subset filtered for scores exceeding 0.95 with a maximum of 50 hits per gene.
Are underlying expression levels included?
Yes, the repository includes estimated expression matrices by RNA-seq run along with corresponding run metadata.
What supplementary network topologies are provided alongside co-expression?
The collection includes a BlastP protein sequence similarity network and a co-evolution (CoEvo) network.
Written from CryptoCEN's own dataset card. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.