MicroRNA (miRNA) are short, non-coding RNAs involved in cell regulation at posttranscriptional and translational levels. MiRNA act on messenger RNA through a silencing mechanism, affecting biological activities such as cell cycle control, biological development, differentiation, and stress response. Experimental validation of predicted miRNA is a time and cost expensive procedure, as it requires wet-lab experiments. Therefore, a variety of computational approaches have been developed to increase prediction accuracy and reduce validation costs. While these methods are highly effective, they require large labelled training data sets, which are often not available for many species. Simultaneously, emerging wet-lab experimental procedures are becoming available that produce large unlabelled data sets of genomic sequence and RNA expression profiles. Existing methods are unable to leverage these unlabelled data. This thesis explores two emerging trends in semi-supervised machine learning to maximize the utility of both labelled and unlabelled training data. Specifically, this thesis explores the application of active learning and multi-view co-training to microRNA prediction for the first time. Results show that our active learning approach is able to greatly improve classification performance using a small number of labeled instances, outperforming state-of-the-art methods under equivalent training data constraints. Multi-view co-training results also demonstrate improved performance compared to single view classifiers and yield high classification performance using a minimum number of labeled instances for classification. This thesis demonstrates that semi-supervised machine learning is likely be useful in creating predictors of novel miRNA, particularly for species where few training exemplars are available.

Tremendous progress has been made over the past few decades to develop skin substitutes for the management of acute and chronic wounds. With the advent of tissue engineering and the ability to combine advanced manufacturing technologies with biomaterials and cell culture systems, more biomimetic tissue constructs have been emerged. Synthetic and natural biomaterials are the main constituents of these skin-like constructs, which play a significant role in tissue grafting, the body's immune response, and the healing process. The act of implanting biomaterials into the human body is subject to the body's immune response, and the complex nature of the immune system involves many different cell types and biological processes that will ultimately determine the success of a skin graft. As such, a large body of recent studies has been focused on the evaluation of the performance and risk assessment of these substitutes. This review summarizes the past and present advances in in vitro, in vivo and clinical applications of tissue-engineered skins. We discuss the role of immunomodulatory biomaterials and biomaterials risk assessment in skin tissue engineering. We will finally offer a roadmap for regulating tissue engineered skin substitutes.

Organ‐on‐chip (OOC) platforms have attracted attentions of pharmaceutical companies as powerful tools for screening of existing drugs and development of new drug candidates. OOCs have primarily used human cell lines or primary cells to develop biomimetic tissue models. However, the ability of human stem cells in unlimited self‐renewal and differentiation into multiple lineages has made them attractive for OOCs. The microfluidic technology has enabled precise control of stem cell differentiation using soluble factors, biophysical cues, and electromagnetic signals. This study discusses different tissue‐ and organ‐on‐chip platforms (i.e., skin, brain, blood–brain barrier, bone marrow, heart, liver, lung, tumor, and vascular), with an emphasis on the critical role of stem cells in the synthesis of complex tissues. This study further recaps the design, fabrication, high‐throughput performance, and improved functionality of stem‐cell‐based OOCs, technical challenges, obstacles against implementing their potential applications, and future perspectives related to different experimental platforms.

Methods for the de novo identification of microRNA (miRNA) have been developed using a range of sequence-based features. With the increasing availability of next generation sequencing (NGS) transcriptome data, there is a need for miRNA identification that integrates both NGS transcript expression-based patterns as well as advanced genomic sequence-based methods. While miRDeep2 does examine the predicted secondary structure of putative miRNA sequences, it does not leverage many of the sequence-based features used in state-of-the-art de novo methods. Meanwhile, other NGS-based methods, such as miRanalyzer, place an emphasis on sequence-based features without leveraging advanced expression-based features reflecting miRNA biosynthesis. This represents an opportunity to combine the strengths of NGS-based analysis with recent advances in de novo sequence-based miRNA prediction. We here develop a method, microRNA Prediction using Integrated Evidence (miPIE), which integrates both expression-based and sequence-based features to achieve significantly improved miRNA prediction performance. Feature selection identifies the 20 most discriminative features, 3 of which reflect strictly expression-based information. Evaluation using precision-recall curves, for six NGS data sets representing six diverse species, demonstrates substantial improvements in prediction performance compared to three methods: miRDeep2, miRanalyzer, and mirnovo. The individual contributions of expression-based and sequence-based features are also examined and we demonstrate that their combination is more effective than either alone.

MicroRNA (miRNA) are short, non-coding RNAs involved in cell regulation at post-transcriptional and translational levels. Numerous computational predictors of miRNA been developed that generally classify miRNA based on either sequence- or expression-based features. While these methods are highly effective, they require large labelled training data sets, which are often not available for many species. Simultaneously, emerging high-throughput wet-lab experimental procedures are producing large unlabelled data sets of genomic sequence and RNA expression profiles. Existing methods use supervised machine learning and are therefore unable to leverage these unlabelled data. In this paper, we design and develop a multi-view co-training approach for the classification of miRNA to maximize the utility of unlabelled training data by taking advantage of multiple views of the problem. Starting with only 10 labelled training data, co-training is shown to significantly (p < 0.01) increase classification accuracy of both sequence- and expression-based classifiers, without requiring any new labelled training data. After 11 iterations of co-training, the expression-based view of miRNA classification experiences an average increase in AUPRC of 15.81% over six species, compared to 11.90% for self-training and 4.84% for passive learning. Similar results are observed for sequence-based classifiers with increases of 46.47%, 39.53% and 29.43%, for co-training, self-training, and passive learning, respectively. The final co-trained sequence and expression-based classifiers are integrated into a final confidence-based classifier which shows improved performance compared to both the expression (1.5%, p = 0.021) and sequence (3.7%, p = 0.006) views. This study represents the first application of multi-view co-training to miRNA prediction and shows great promise, particularly for understudied species with few available training data.