Analysis of Human Coding Genes Reveals Unsettled Status for Many
Author Information
Author(s): Maquedano Miguel, Cerdán-Vélez Daniel, Tress Michael L.
Primary Institution: Bioinformatics Unit, Spanish National Cancer Research Centre (CNIO)
Hypothesis
Are the current estimates of human coding genes accurate given the discrepancies among major gene annotation databases?
Conclusion
The study found that there are still 2,606 annotated genes with unresolved coding status among the three major gene databases.
Supporting Evidence
- Previous analyses showed that many annotated coding genes were later reclassified as non-coding.
- Discrepancies among the three databases indicate ongoing challenges in gene annotation.
- More than 700 genes have been merged or reclassified since the last analysis.
Takeaway
Scientists looked at human genes and found that many of them might not actually make proteins, even though they are labeled as coding genes.
Methodology
The study involved merging and analyzing coding genes from three major databases: Ensembl/GENCODE, RefSeq, and UniProtKB.
Potential Biases
Potential bias exists due to reliance on annotations that may overestimate the number of coding genes.
Limitations
The analysis may not account for all potential non-coding features and relies on existing annotations that may be inaccurate.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website