Entropy-based model for miRNA isoform analysis

GeneAI 3.0: powerful, novel, generalized hybrid and ensemble deep learning frameworks for miRNA species classification of stationary patterns from nucleotides

Due to the intricate relationship between the small non-coding ribonucleic acid (miRNA) sequences, the classification of miRNA species, namely Human, Gorilla, Rat, and Mouse is challenging. Previous methods are not robust and accurate. In this study, we present AtheroPoint's GeneAI 3.0, a powerful, novel, and generalized method for extracting features from the fixed patterns of purines and pyrimidines in each miRNA sequence in ensemble paradigms in machine learning (EML) and convolutional neural network (CNN)-based deep learning (EDL) frameworks. GeneAI 3.0 utilized five conventional (Entropy, Dissimilarity, Energy, Homogeneity, and Contrast), and three contemporary (Shannon entropy, Hurst exponent, Fractal dimension) features, to generate a composite feature set from given miRNA sequences which were then passed into our ML and DL classification framework. A set of 11 new classifiers was designed consisting of 5 EML and 6 EDL for binary/multiclass classification. It was benchmarked against 9 solo ML (SML), 6 solo DL (SDL), 12 hybrid DL (HDL) models, resulting in a total of 11 + 27 = 38 models were designed. Four hypotheses were formulated and validated using explainable AI (XAI) as well as reliability/statistical tests. The order of the mean performance using accuracy (ACC)/area-under-the-curve (AUC) of the 24 DL classifiers was: EDL > HDL > SDL. The mean performance of EDL models with CNN layers was superior to that without CNN layers by 0.73%/0.92%. Mean performance of EML models was superior to SML models with improvements of ACC/AUC by 6.24%/6.46%. EDL models performed significantly better than EML models, with a mean increase in ACC/AUC of 7.09%/6.96%. The GeneAI 3.0 tool produced expected XAI feature plots, and the statistical tests showed significant p-values. Ensemble models with composite features are highly effective and generalized models for effectively classifying miRNA sequences.

Tumor classification and biomarker discovery based on the 5'isomiR expression level

Background

The miRNA isoforms (isomiRs) have been suggested to regulate the same pathways as the canonical miRNA and play an important biological role in miRNA-mediated gene regulation. Recently, a study has demonstrated that the presence or absence of all isomiRs could efficiently discriminate amongst 32 TCGA cancer types. Besides, an effective reduction of distinguishing isomiR features for multiclass tumor discrimination must have a major impact on our understanding of the disease and treatment of cancer.

Methods

In this study, we have constructed a combination of the genetic algorithms (GA) with Random Forest (RF) algorithms to detect reliable sets of cancer-associated 5’isomiRs from TCGA isomiR expression data for multiclass tumor classification.

Results

We obtained 100 sets of the optimal predictive features, each of which comprised of 50–5’isomiRs that could effectively classify with an average sensitivity of 92% samples from 32 different tumor types. We calculated the frequency with which a 5’isomiR found in these sets as measuring its importance for tumor classification. Many highly frequent 5’isomiRs with different 5′ loci from canonical miRNAs were detected in these sets, supporting that the isomiRs play a significant role in the multiclass tumor classification. The further functional enrichment analysis showed that the target genes of the 10 most frequently appearing 5’isomiRs were involved in the activity of transcription activator and protein kinase and cell-cell adhesion.

Conclusions

The findings of the present study indicated that the 5’isomiRs might be employed for multiclass tumor classification and the suggested that GA/RF model could perform effective tumor classification by a series of largely independent optimal predictor 5′ isomiR sets.

Dysregulation of miRNA isoform level at 5' end in Alzheimer's disease

Alzheimer's disease (AD) is the most common form of dementia, whose mechanism is still not yet fully understood. A miRNA-based signature method, commonly according to the changes of expression levels, is widely used for AD analysis in previous studies. Recently, miRNA isoforms called as isomiR variants, which is considered to play important biological roles, have been demonstrated as the applications of high throughput sequencing platforms. Here, we presented an entropy-based model to detect the miRNA isoform level at the 5' end, and found many miRNAs with significant changes of isoform levels between the early stage and the late stage of AD by the application of this model to the public data. The statistical significance of the overlap between isoform-level changed miRNAs and AD related miRNAs extracted from HMDD2 supports that these miRNA isoforms are not degradation products. Based on the most common isomiR seed analysis of isoform-level changed AD related miRNAs, the predicted targets are also found to be enriched for genes involved in transcriptional regulation and the nervous system. After comparing with the expression level based method, we detected that changes of 5' isoform levels are more stable than those of expression levels for AD related miRNA detecting.

miRge - A Multiplexed Method of Processing Small RNA-Seq Data to Determine MicroRNA Entropy

Small RNA RNA-seq for microRNAs (miRNAs) is a rapidly developing field where opportunities still exist to create better bioinformatics tools to process these large datasets and generate new, useful analyses. We built miRge to be a fast, smart small RNA-seq solution to process samples in a highly multiplexed fashion. miRge employs a Bayesian alignment approach, whereby reads are sequentially aligned against customized mature miRNA, hairpin miRNA, noncoding RNA and mRNA sequence libraries. miRNAs are summarized at the level of raw reads in addition to reads per million (RPM). Reads for all other RNA species (tRNA, rRNA, snoRNA, mRNA) are provided, which is useful for identifying potential contaminants and optimizing small RNA purification strategies. miRge was designed to optimally identify miRNA isomiRs and employs an entropy based statistical measurement to identify differential production of isomiRs. This allowed us to identify decreasing entropy in isomiRs as stem cells mature into retinal pigment epithelial cells. Conversely, we show that pancreatic tumor miRNAs have similar entropy to matched normal pancreatic tissues. In a head-to-head comparison with other miRNA analysis tools (miRExpress 2.0, sRNAbench, omiRAs, miRDeep2, Chimira, UEA small RNA Workbench), miRge was faster (4 to 32-fold) and was among the top-two methods in maximally aligning miRNAs reads per sample. Moreover, miRge has no inherent limits to its multiplexing. miRge was capable of simultaneously analyzing 100 small RNA-Seq samples in 52 minutes, providing an integrated analysis of miRNA expression across all samples. As miRge was designed for analysis of single as well as multiple samples, miRge is an ideal tool for high and low-throughput users. miRge is freely available at http://atlas.pathology.jhu.edu/baras/miRge.html.