Vol 19, No 1 (2024)

Over the past two decades, studies have discovered a special form of alternative splicing (AS) that produces a circular form of RNA. This stands in contrast to normal AS, which produces a linear form of RNA. Although these circRNAs have garnered considerable attention in the scientific community for their biogenesis and functions, the focus of these studies has been on the regulatory role of circRNAs with the assumption that circRNAs are non-coding. As non-coding RNAs, they may regulate mRNA transcription, tumor initiation, and translation by sponging miRNAs and RNA-binding proteins (RBPs). In addition to these regulatory roles of circRNAs, however, recent studies have provided strong evidence for their translation. The translation of circRNAs is expected to have an important role in promoting cancer cell growth and activating molecular pathways related to cancer development. In some cases, the translation of circRNAs is shown to be efficiently driven by an internal ribosome entry site (IRES). The development of a computational tool for identifying and characterizing the translation of circRNAs using high-throughput sequencing and IRES increases identifiable proteins translated from circRNAs. In turn, it has a substantial impact on helping researchers understand the functional role of proteins derived from circRNAs. New web resources for aggregating, cataloging, and visualizing translational information of circRNAs derived from previous studies have been developed. In this paper, general concepts of circRNA, circRNA biogenesis, translation of circRNA, and existing circRNA tools and databases are summarized to provide new insight into circRNA studies.

Treatment of many human diseases involves small-molecule drugs.Some target proteins, however, are not druggable with traditional strategies. Innovative RNA-targeted therapeutics may overcome such a challenge. Long noncoding RNAs (lncRNAs) are transcribed RNAs that do not translate into proteins. Their ability to interact with DNA, RNA, microRNAs (miRNAs), and proteins makes them an interesting target for regulating gene expression and signaling pathways.In the past decade, a catalog of lncRNAs has been studied in several human diseases. One of the challenges with lncRNA studies include their lack of coding potential, making, it difficult to characterize them in wet-lab experiments functionally. Several computational tools have thus been designed to characterize functions of lncRNAs centered around lncRNA interaction with proteins and RNA, especially miRNAs. This review comprehensively summarizes the methods and tools for lncRNA-RNA interactions and lncRNA-protein interaction prediction.We discuss the tools related to lncRNA interaction prediction using commonlyused models: ensemble-based, machine-learning-based, molecular-docking and network-based computational models. In biology, two or more genes co-expressed tend to have similar functions. Coexpression network analysis is, therefore, one of the most widely-used methods for understanding the function of lncRNAs. A major focus of our study is to compile literature related to the functional prediction of lncRNAs in human diseases using co-expression network analysis. In summary, this article provides relevant information on the use of appropriate computational tools for the functional characterization of lncRNAs that help wet-lab researchers design mechanistic and functional experiments.

Objective:Representation learning in the context of biological concepts involves acquiring their numerical representations through various sources of biological information, such as sequences, interactions, and literature. This study has conducted a comprehensive systematic review by analyzing both quantitative and qualitative data to provide an overview of this field.

Methods:Our systematic review involved searching for articles on the representation learning of biological concepts in PubMed and EMBASE databases. Among the 507 articles published between 2015 and 2022, we carefully screened and selected 65 papers for inclusion. We then developed a structured workflow that involved identifying relevant biological concepts and data types, reviewing various representation learning techniques, and evaluating downstream applications for assessing the quality of the learned representations.

Results:The primary focus of this review was on the development of numerical representations for gene/DNA/RNA entities. We have found Word2Vec to be the most commonly used method for biological representation learning. Moreover, several studies are increasingly utilizing state-of-the-art large language models to learn numerical representations of biological concepts. We also observed that representations learned from specific sources were typically used for single downstream applications that were relevant to the source.

Conclusion:Existing methods for biological representation learning are primarily focused on learning representations from a single data type, with the output being fed into predictive models for downstream applications. Although there have been some studies that have explored the use of multiple data types to improve the performance of learned representations, such research is still relatively scarce. In this systematic review, we have provided a summary of the data types, models, and downstream applications used in this task.

Background:Root is a desirable trait for modern plant breeding programs, as the roots play a pivotal role in the growth and development of plants. Therefore, identification of the genes governing the root traits is an essential research component. With regard to the identification of root-associated genes/proteins, the existing wet-lab experiments are resource intensive and the gene expression studies are species-specific. Thus, we proposed a supervised learning-based computational method for the identification of root-associated proteins.

Method:The problem was formulated as a binary classification, where the root-associated proteins and non-root-associated proteins constituted the two classes. Four different machine learning algorithms such as support vector machine (SVM), extreme gradient boosting, random forest, and adaptive boosting were employed for the classification of proteins of the two classes. Sequence-derived features such as AAC, DPC, CTD, PAAC, and ACF were used as input for the learning algorithms.

Results:The SVM achieved higher accuracy with the 250 selected features of AAC+DPC+CTD than that of other possible combinations of feature sets and learning algorithms. Specifically, SVM with the selected features achieved overall accuracies of 0.74, 0.73, and 0.73 when evaluated with single 5-fold cross-validation (5F-CV), repeated 5F-CV, and independent test set, respectively.

Conclusions:A web-enabled prediction tool SVM-Root (https://iasri-sg.icar.gov.in/svmroot/) has been developed for the computational prediction of the root-associated proteins. Being the first of its kind, the proposed model is believed to supplement the existing experimental methods and high throughput GWAS and transcriptome studies.