The state of play in higher eukaryote gene annotation

Abstract

A genome sequence is worthless if it cannot be deciphered, therefore efforts to describe — or ‘annotate’ — genes began as soon as DNA sequences became available. Whereas early work focused on individual protein-coding genes, the modern genomic ocean is a complex maelstrom of alternative splicing, non-coding transcription and pseudogenes. Scientists — from clinicians to evolutionary biologists — need to navigate these waters, and this has led to the design of high-throughput, computationally driven annotation projects. The catalogues produced are key resources for genome exploration, especially as they become integrated with expression, epigenomic and variation datasets. Their creation, however, remains challenging.

Recognition of the polycistronic nature of human genes is critical to understanding the genotype-phenotype relationship

“…Different strategies have been adopted over the past years, which essentially regroup two goals: (1) to identify transcript structure (e.g., intron vs. exon); and (2) to identify the functional potential (e.g., contains a CDS) (Pruitt et al 2009;Harrow et al 2012;Aken et al 2016;Mudge and Harrow 2016). These pipelines, however, invoke a uni-coding presumption.…”

“…This means that despite the large number of smORF and alternative ORF discoveries, only a limited number make it through to genome annotation (Southan 2017). The current genome annotation system has been blamed for simplifying a transcript's definition, not taking into account their potential to hold multiple functional features (for review, see Mudge and Harrow 2016).…”

“…However, because false positives could burden variant-calling workflows, putative functional annotations are removed at the filtration and curation steps (Quality Control on Fig. 4; Koonin and Galperin 2003;Mudge and Harrow 2016). However, as discussed earlier, genome specificity needs might differ based on the experimental purpose (variant calling, novel protein identification, etc.).…”

“…As shown in Figure 4, most pipelines annotate ORFs and subsequent protein products from ab initio ORF prediction or sequence alignment with known proteins (from the UniProt or RefSeq databases) (Keller et al 2011; The UniProt Consortium 2014). ORF prediction mostly relies on ORF size, codon usage, and the nonsynonymous to synonymous mutation ratio (Pruitt et al 2009;Keller et al 2011;Mudge and Harrow 2016). This means that current genome annotations are shaped by evolution-, prior knowledge-, and hypothesis-driven data.…”

Biological Pathway Specificity in the Cell—Does Molecular Diversity Matter?

“…The human mind is bound to think in linear cause‐and‐effect paradigms since many events in our everyday experience appear to be the result of identifiable causes. Consequently, the basic assignment of functions to the many nonprotein coding (or noncoding) RNAs discovered to be encoded in our genome, for example, is seen as tied to the unresolved question of whether or not they participate in any specific intracellular pathway, with the expectation that their removal should lead to identifiable phenotypic changes . Similarly, bioscience textbooks teach that the specificity of, say, an enzyme is a direct consequence of its molecular structure in relation to only that of its specific substrate; and that a metabolic pathway is a very specific sequence of reactions linked through a linear flux of interrelated substrates.…”