Give GO annotation a go – Part III

In the previous blog about GO annotation we discussed how GO terms are assigned only based on supporting data in the paper. In this blog we will go into further detail for a specific type of supporting data: phenotypic data.

Phenotypic data are mainly collected for mutants, either T-DNA insertion lines or mutants created through methods such as CRISPR or chemical mutagenesis. Based on the phenotype of the mutant, inferences are made about the function of the gene. Curators at TAIR capture this information as mutant phenotypes. However, we do not usually capture this information as GO annotations when such phenotypes represent the indirect effects of the mutated gene.

For example, the RRP7 (AT5G38720) mutants described by Micol-Ponce et al. (2018) have abnormal leaf shape and phyllotactic patterning. However, since RRP7 plays a role in ribosomal RNA maturation, a very general process and is clearly a ‘housekeeping’ gene, these mutant phenotypes were not captured as GO annotations

Open

but rather as phenotypes of the germplasm.

Open

In other cases, a mutant phenotype represents a more direct effect allowing for the assignment of a GO annotation. For example, the emb3103 (aka pdm2-1) (AT1G10910) mutant described by Du et al. (2017) showed an effect on chloroplast RNA editing, which has a direct effect on making a functional chloroplast. As a result, the GO annotations “chloroplast RNA modification” and "chloroplast organization" were assigned based on mutant phenotype evidence.

Open

The phenotype is also captured and linked to the germplasm record.

Open

Hopefully this is helpful and will make you feel like contributing to the GO annotation effort by submitting your paper’s GO info to TOAST (toast.arabidopsis.org).