XSAnno: a framework for building ortholog models in cross-species transcriptome comparisons
Ying Zhu, Mingfeng Li, André M. M. Sousa, Nenad Šestan
Abstract
Open-access reader
Ying Zhu, Mingfeng Li, André M. M. Sousa, Nenad Šestan
Abstract
Open-access reader
BACKGROUND: The accurate characterization of RNA transcripts and expression levels across species is critical for understanding transcriptome evolution. As available RNA-seq data accumulate rapidly, there is a great demand for tools that build gene annotations for cross-species RNA-seq analysis. However, prevailing methods of ortholog annotation for RNA-seq analysis between closely-related species do not take inter-species variation in mappability into consideration. RESULTS: Here we present XSAnno, a computational framework that integrates previous approaches with multiple filters to improve the accuracy of inter-species transcriptome comparisons. The implementation of this approach in comparing RNA-seq data of human, chimpanzee, and rhesus macaque brain transcriptomes has reduced the false discovery of differentially expressed genes, while maintaining a low false negative rate. CONCLUSION: The present study demonstrates the utility of the XSAnno pipeline in building ortholog annotations and improving the accuracy of cross-species transcriptome comparisons.
OpenAlex reports 36 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
BACKGROUND: The accurate characterization of RNA transcripts and expression levels across species is critical for understanding transcriptome evolution. As available RNA-seq data accumulate rapidly, there is a great demand for tools that build gene annotations for cross-species RNA-seq analysis. However, prevailing methods of ortholog annotation for RNA-seq analysis between closely-related species do not take inter-species variation in mappability into consideration. RESULTS: Here we present XSAnno, a computational framework that integrates previous approaches with multiple filters to improve the accuracy of inter-species transcriptome comparisons. The implementation of this approach in comparing RNA-seq data of human, chimpanzee, and rhesus macaque brain transcriptomes has reduced the false discovery of differentially expressed genes, while maintaining a low false negative rate. CONCLUSION: The present study demonstrates the utility of the XSAnno pipeline in building ortholog annotations and improving the accuracy of cross-species transcriptome comparisons.
Key concepts: Biology, Transcriptome, Computational biology, DNA microarray, Evolutionary biology, Proteomics, Genetics, Gene