Understanding the Causes of Errors in Eukaryotic Protein-coding Gene Prediction: A Case Study of Primate Proteomes

Abstract

Background. Recent advances in sequencing technologies have led to an explosion in the number of genomes available, but accurate genome annotation remains a major challenge. The prediction of protein-coding genes in eukaryotic genomes is especially problematic, due to their complex exon-intron structures. Even the best eukaryotic gene prediction algorithms can make serious errors that will significantly affect subsequent analyses. Results. We first investigated the prevalence of gene prediction errors in a large set of 176,478 proteins from ten primate proteomes available in public databases. Using the well-studied human proteins as a reference, a total of 82,305 potential errors were detected, including 44,001 deletions, 27,289 insertions and 11,015 mismatched segments where part of the correct protein sequence is replaced with an alternative erroneous sequence. We then focused on the mismatched sequence errors that cause particular problems for downstream applications. A detailed characterization allowed us to identify the potential causes for the gene misprediction in approximately half (5446) of these cases. As a proof-of-concept, we also developed a simple method which allowed us to propose improved sequences for 603 primate proteins. Conclusions. Gene prediction errors in primate proteomes affect up to 50% of the sequences. Major causes of errors include undetermined genome regions, genome sequencing or assembly issues, and limitations in the models used to represent gene exon-intron structures. Nevertheless, existing genome sequences can still be exploited to improve protein sequence quality. Perspectives of the work include the characterization of other types of gene prediction errors, as well as the development of a more comprehensive algorithm for protein sequence error correction.

Spliceator: multi-species splice site prediction using convolutional neural networks

“…Genscan [7], Snap [8] or GlimmerHMM [9]. Despite these developments, the annotation of gene structure remains a major challenge, especially for eukaryotic organisms [10][11][12][13] due to their complex exon-intron mosaics [14] (Fig. 1).…”

CeGAL: revisiting a widespread fungal-specific TF family using an in silico error-aware approach to identify missing zinc cluster domains

“…Some important causes of erroneous sequences have been identified, including the genome sequence quality and gene structure complexity (40), as well as redundant or conflicting information in different resources or in the literature (34,41). Consequently, it has been estimated that 40 to 60% of the protein sequences in public databases are erroneous (42)(43)(44).…”

“…Typical errors include missing exons, non-coding sequence retention in exons, wrong exon and gene boundaries, fragmenting genes and merging neighboring genes. Thus, the development of automated methods to identify and correct mispredicted protein sequences remains an important research topic (43, 45–47). Our study showing high error rates at a family-wide level is further evidence of the potential of domain-centric approaches for sequence annotation correction and of the urgent need to clean the databases.…”

MERLIN: Identifying Inaccuracies in Multiple Sequence Alignments Using Object Detection

“…In order to train and evaluate our object detector, we used multiple sequence alignments from an in-house built dataset [15]. These MSAs are extracted from the Uniprot reference proteomes [5] and RefSeq [17] databases, and were automatically annotated using SIBIS algorithm [12].…”

“…Algorithms used to predict protein-coding genes in DNA sequences, for instance, are not always accurate, and often lead to sequence prediction errors. Consequently, today's protein databases are riddled with inconsistencies [15].…”