Fig. 2
From: Biological factors in the synthetic construction of overlapping genes
Left: Proportions of the dataset used in Opuu et al. with different match identities in SWISS-PROT - virus genes from this dataset have a higher average identity to a SWISS-PROT entry than non-virus genes. Right: Percentage of functional OLGs for the original dataset from Opuu et al. and the average of 10 curated datasets grouped into virus and non virus genes. In curated datasets all original sequences have an exact match in SWISS-PROT. Each curated dataset has 100 sequences with 70-100 amino acids. The virus versus non-virus difference observed in the previous study’s dataset vanishes for the curated datasets