Last rolls of the yoyo: Assessing the human canonical protein count

Abstract

In 2004, when the protein estimate from the finished human genome was only 24,000, the surprise was compounded as reviewed estimates fell to 19,000 by 2014. However, variability in the total canonical protein counts (i.e. excluding alternative splice forms) of open reading frames (ORFs) in different annotation portals persists. This work assesses these differences and possible causes. A 16-year analysis of Ensembl and UniProtKB/Swiss-Prot shows convergence to a protein number of ~20,000. The former had shown some yo-yoing, but both have now plateaued. Nine major annotation portals, reviewed at the beginning of 2017, gave a spread of counts from 21,819 down to 18,891. The 4-way cross-reference concordance (within UniProt) between Ensembl, Swiss-Prot, Entrez Gene and the Human Gene Nomenclature Committee (HGNC) drops to 18,690, indicating methodological differences in protein definitions and experimental existence support between sources. The Swiss-Prot and neXtProt evidence criteria include mass spectrometry peptide verification and also cross-references for antibody detection from the Human Protein Atlas. Notwithstanding, hundreds of Swiss-Prot entries are classified as non-coding biotypes by HGNC. The only inference that protein numbers might still rise comes from numerous reports of small ORF (smORF) discovery. However, while there have been recent cases of protein verifications from previous miss-annotation of non-coding RNA, very few have passed the Swiss-Prot curation and genome annotation thresholds. The post-genomic era has seen both advances in data generation and improvements in the human reference assembly. Notwithstanding, current numbers, while persistently discordant, show that the earlier yo-yoing has largely ceased. Given the importance to biology and biomedicine of defining the canonical human proteome, the task will need more collaborative inter-source curation combined with broader and deeper experimental confirmation in vivo and in vitro of proteins predicted in silico. The eventual closure could be well be below ~19,000.

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

“…As a result, the foreseen increase in smORF count in Swiss-Prot falls short, with an increase from 3.1% in 2009 to 3.3% in 2017 (Southan 2017). 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).…”

“…ORF-prediction algorithms apply the criteria of a single CDS per transcript, and a minimum length of 100 codons, unless the sequence bears high similarity to known proteins or domains (Furuno et al 2003;Pruitt et al 2012;Aken et al 2016). As a result, the foreseen increase in smORF count in Swiss-Prot falls short, with an increase from 3.1% in 2009 to 3.3% in 2017 (Southan 2017). 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).…”

“…4). Such an annotation pipeline would prevent some of today's pitfalls, abolishing the unique CDS presumption and empowering experimental data as well as conservation signatures Southan 2017). This would add a layer of experiment-driven data to genome annotation pipelines.…”

The Role of Long Non-coding RNAs in Human Imprinting Disorders: Prospective Therapeutic Targets

“…The number of lncRNA genes in the human genome has been estimated at 20,000 to 100,000 (Zhao et al, 2016;Fang et al, 2018;Uszczynska-Ratajczak et al, 2018). This number is greater than the canonical protein-coding genes in the human genome (Southan, 2017). lncRNAs are primarily retained in the nucleus, having short half-lives and a rapid turnover rate compared to mRNAs (Clark et al, 2012;Derrien et al, 2012;Yoon et al, 2015).…”

GenOrigin: A Comprehensive Protein-coding Gene Origination Database on the Evolutionary Timescale of Life

“…(42), their gene annotation is still requiring improvement, especially for non-model species. Also, the gene age inferred by GenOrigin is based on the orthology information of Ensembl Compara(14,15), which perform great for old genes but not for new genes.…”