Characterizing ncRNAs in human pathogenic protists using high-throughput sequencing technology

Abstract

ncRNAs are key genes in many human diseases including cancer and viral infection, as well as providing critical functions in pathogenic organisms such as fungi, bacteria, viruses, and protists. Until now the identification and characterization of ncRNAs associated with disease has been slow or inaccurate requiring many years of testing to understand complicated RNA and protein gene relationships. High-throughput sequencing now offers the opportunity to characterize miRNAs, siRNAs, small nucleolar RNAs (snoRNAs), and long ncRNAs on a genomic scale, making it faster and easier to clarify how these ncRNAs contribute to the disease state. However, this technology is still relatively new, and ncRNA discovery is not an application of high priority for streamlined bioinformatics. Here we summarize background concepts and practical approaches for ncRNA analysis using high-throughput sequencing, and how it relates to understanding human disease. As a case study, we focus on the parasitic protists Giardia lamblia and Trichomonas vaginalis, where large evolutionary distance has meant difficulties in comparing ncRNAs with those from model eukaryotes. A combination of biological, computational, and sequencing approaches has enabled easier classification of ncRNA classes such as snoRNAs, but has also aided the identification of novel classes. It is hoped that a higher level of understanding of ncRNA expression and interaction may aid in the development of less harsh treatment for protist-based diseases.

Do miRNAs have a deep evolutionary history?

“…We set out to test an intriguing and revolutionary hypothesis, that miRNAs evolved once in the last common ancestor (LCA) of crown eukaryotes and that this ancestor shared miRNAs present today in plants and animals. Although this hypothesis is becoming widely accepted 77, 78, 88–91 our data clearly refute it. Instead, our results show that of the 73 plant and animal miRNAs identified in protists, all fail to meet the criteria required for the identification of miRNAs.…”

“…Some of the studies 27–30 describing miRNAs in protists did not follow the established conventions for naming miRNAs 16, 62, 63. Consequently, authors annotated novel miRNAs using names of already existing animal and plant miRNAs, implying homology, which can readily be misinterpreted 77, 78. For example, Saraiya et al 28 described miR‐2 regulating the expression of 22 variant surface protein genes in Giardia .…”

The parasite Trichomonas vaginalis expresses thousands of pseudogenes and long non-coding RNAs independently from functional neighbouring genes

“…Long non-coding RNAs (lncRNAs) are often defined as transcribed but not translated RNA segments larger than sRNAs (>200 nucleotides) [ 29 ]. lncRNAs affect chromosomal dynamics, the telomeres and structural organization [ 20 , 21 , 23 ].…”

Identificação de snoRNAs usando aprendizagem de máquina

“…Métodos computacionais para a identificação e classificação de ncRNAs vêm sendo propostos e aprimorados nos últimos anos [63,34,5,95,13]. A tarefa de identificar e classificar ncRNAs é bastante desafiadora, devido a dificuldade de confirmar experimentalmente a função de um ncRNA, pois essa está associada à sua estrutura espacial (estruturas secundária e terciária), o que impede o uso de métodos de predição de genes codificadores de proteínas que usam apenas a informação de sua estrutura primária (sequência de nucleotídeos).…”

“…These computational methods focus on predicting candidates that have to be experimentally confirmed. Identification of ncRNAs have been developed for a variety of organisms [31,13,96,53], with the objective of constructing sets of different classes of ncRNAs. In particular, snoR-NAs [25] are 60 to 300 nt ncRNAs, classified based on their characteristic sequence elements, called boxes, in two main classes: H/ACA box snoRNAs and C/D box snoRNAs.…”