iPiDA-LTR: Identifying piwi-interacting RNA-disease associations based on Learning to Rank

Abstract

Piwi-interacting RNAs (piRNAs) are regarded as drug targets and biomarkers for the diagnosis and therapy of diseases. However, biological experiments cost substantial time and resources, and the existing computational methods only focus on identifying missing associations between known piRNAs and diseases. With the fast development of biological experiments, more and more piRNAs are detected. Therefore, the identification of piRNA-disease associations of newly detected piRNAs has significant theoretical value and practical significance on pathogenesis of diseases. In this study, the iPiDA-LTR predictor is proposed to identify associations between piRNAs and diseases based on Learning to Rank. The iPiDA-LTR predictor not only identifies the missing associations between known piRNAs and diseases, but also detects diseases associated with newly detected piRNAs. Experimental results demonstrate that iPiDA-LTR effectively predicts piRNA-disease associations outperforming the other related methods.

Databases and computational methods for the identification of piRNA-related molecules: A survey

“…In addition, the one-hot encoding method may lead to high-dimensional sparse feature representation, which increases the computational complexity. Cardiovascular disease,Renal cell carcinoma and Alzheimer’s disease 13 iPiDA-LTR [33] SVM&LR&RF&CF Disease semantic similarity and piRNA sequence similarity 4350 21 5002 0.9543 Using learning ranking method, transforming the prediction problem into a search task, and modeling the piRNA-disease association from a global perspective. The performance of learning ranking method also depends on the selection and evaluation of ranking metrics.…”

“…The model exhibits high performance and robustness on a benchmark dataset and a new dataset. Zhang et al [33] proposed an approach to identify piRNA–disease association, following the idea of learning to rank (LTR) [148] , [149] proposed by Wei et al [150] . Specifically, such an identification or classification problem is formalized as a search task, where the target piRNA and diseases are viewed as the query and the document candidates, respectively, and the association between the piRNA and diseases is positively correlated to the ranking position of the disease.…”

“…For example, SPRDA [147] uses a heterogeneous network, but the structural perturbation method of this network may introduce noise and error; (2) the need to consider the association strength between piRNAs and diseases, rather than just binary association, which may require more fine-grained evaluation metrics and ranking methods. For example, PDA-LTR [33] used a ranking learning method, but determining the ranking and loss functions, as well as dealing with the problem of different lengths of queries and documents were required in this method.…”

“…iPiDi-PUL [22] predicted that piR-hsa-1849 was associated with breast cancer, and it was downregulated in breast cancer tissues and may be used as a potential biomarker for breast cancer. iPiDA-LTR [33] predicted that piR-hsa-23210 and piR-hsa-23209 were associated with multiple diseases, and their target genes played important roles in human spermatogenesis and nervous system development and may be key therapeutic targets. iPiDA-GCN [34] predicted that piR-hsa-31280 and piR-hsa-8245 were associated with cardiovascular diseases, and they were abnormally expressed in cardiovascular disease tissues and may be related to cardiac function and repair.…”

ETGPDA: identification of piRNA-disease associations based on embedding transformation graph convolutional network

“…The impact of negative samples was eliminated and it increased the structure consistency index to measure the feasibility of prediction, which has achieved high prediction performance. Zhang et al [ 17 ] developed a model to identify the PDAs (iPiDA-LTR), which was based on learning sequencing. The iPiDA-LTR can not only identify the deletion associations between known piRNA and diseases but also detect the associations with potential PDAs.…”