Dr Gloria Despacio-Reyes

Senior Computer Biologist


This person is a member of Sanger Institute Alumni.

I am with the vertebrate annotation group working on genome analysis and annotation of coding, noncoding genes and pseudogenes, including their genomic features. As part of the team, I was involved in the generation of human GENCODE geneset of the ENCODE Project. We are currently annotating the mouse genome. Coming from a broad science background, my interest is on annotating non-coding genes, which is challenging and not a straighforward process. More manual intervention and careful assessment of available experimental evidences are very crucial. Currently, the use of NGS, utilizing long reads of transcriptomic data and RNA-seq models, are showing so much potential in mapping, aligning and annotating non-coding genes.

I currently worked on using long reads of RNA-seq models to improve the annotation of non-coding genes, or to create novel ones. I am happy to have taken an active part in the annotation of the long non-coding genes, which has progressed tremendously over the years. I worked meticulously and I have been involved in a number of mini-projects of the team, such as in verifying non-coding genes tested experimentally by our partners.

The absence of translation or protein product in non-coding genes makes it tricky to annotate, if not difficult. For example, since there is no methionine, which usually marks the canonical transcription start site (TSS), we rely on some genomic feature alignment such as CAGE (cap analysis of gene expression) tags and other software-predicted TSS to support transcription, together with the correct splicing junctions. We are now starting to explore the use of NGS technologies to help us improve and fast track, most specially on annotating the non-coding genes.

I am passionate about my work. For one thing, HAVANA genome annotation is strongly experimentally evidence-based. This keeps me very much in touch with the laboratory world (which was what I used to do), while doing computer work on my desk. In theory, our annotation should be error-free, as the datasets we generate are used by the larger science community. Any erroneous data might proliferate uncontrollably, with annotation that is incorrect, or not thorough and comprehensive. Hence, the job of an annotator is not an easy one. Our robust system- the software, the annotation rules and guidelines and the QC and consistency check- all these contribute so we are able to produce such ‘gold standard genome annotation’ that is now widely used and adopted. Being able to help polish genome information, refine it, revise, update or attach expert remarks – and all things that a genome annotator can do for science, is a great task I am proud of.

My timeline


My publications

Loading publications...