Multimodal Biomedical Data Fusion Using Sparse Canonical Correlation Analysis and Cooperative Learning: A Cohort Study on COVID-19.

Through technological innovations, patient cohorts can be examined from multiple views with high-dimensional, multiscale biomedical data to classify clinical phenotypes and predict outcomes. Here, we aim to present our approach for analyzing multimodal data using unsupervised and supervised sparse linear methods in a COVID-19 patient cohort. This prospective cohort study of 149 adult patients was conducted in a tertiary care academic center. First, we used sparse canonical correlation analysis (CCA) to identify and quantify relationships across different data modalities, including viral genome sequencing, imaging, clinical data, and laboratory results. Then, we used cooperative learning to predict the clinical outcome of COVID-19 patients. We show that serum biomarkers representing severe disease and acute phase response correlate with original and wavelet radiomics features in the LLL frequency channel (corr(Xu1, Zv1) = 0.596, p-value < 0.001). Among radiomics features, histogram-based first-order features reporting the skewness, kurtosis, and uniformity have the lowest negative, whereas entropy-related features have the highest positive coefficients. Moreover, unsupervised analysis of clinical data and laboratory results gives insights into distinct clinical phenotypes. Leveraging the availability of global viral genome databases, we demonstrate that the Word2Vec natural language processing model can be used for viral genome encoding. It not only separates major SARS-CoV-2 variants but also allows the preservation of phylogenetic relationships among them. Our quadruple model using Word2Vec encoding achieves better prediction results in the supervised task. The model yields area under the curve (AUC) and accuracy values of 0.87 and 0.77, respectively. Our study illustrates that sparse CCA analysis and cooperative learning are powerful techniques for handling high-dimensional, multimodal data to investigate multivariate associations in unsupervised and supervised tasks.

An automated framework for evaluation of deep learning models for splice site predictions.

A novel framework for the automated evaluation of various deep learning-based splice site detectors is presented. The framework eliminates time-consuming development and experimenting activities for different codebases, architectures, and configurations to obtain the best models for a given RNA splice site dataset. RNA splicing is a cellular process in which pre-mRNAs are processed into mature mRNAs and used to produce multiple mRNA transcripts from a single gene sequence. Since the advancement of sequencing technologies, many splice site variants have been identified and associated with the diseases. So, RNA splice site prediction is essential for gene finding, genome annotation, disease-causing variants, and identification of potential biomarkers. Recently, deep learning models performed highly accurately for classifying genomic signals. Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) and its bidirectional version (BLSTM), Gated Recurrent Unit (GRU), and its bidirectional version (BGRU) are promising models. During genomic data analysis, CNN's locality feature helps where each nucleotide correlates with other bases in its vicinity. In contrast, BLSTM can be trained bidirectionally, allowing sequential data to be processed from forward and reverse directions. Therefore, it can process 1-D encoded genomic data effectively. Even though both methods have been used in the literature, a performance comparison was missing. To compare selected models under similar conditions, we have created a blueprint for a series of networks with five different levels. As a case study, we compared CNN and BLSTM models' learning capabilities as building blocks for RNA splice site prediction in two different datasets. Overall, CNN performed better with [Formula: see text] accuracy ([Formula: see text] improvement), [Formula: see text] F1 score ([Formula: see text] improvement), and [Formula: see text] AUC-PR ([Formula: see text] improvement) in human splice site prediction. Likewise, an outperforming performance with [Formula: see text] accuracy ([Formula: see text] improvement), [Formula: see text] F1 score ([Formula: see text] improvement), and [Formula: see text] AUC-PR ([Formula: see text] improvement) is achieved in C. elegans splice site prediction. Overall, our results showed that CNN learns faster than BLSTM and BGRU. Moreover, CNN performs better at extracting sequence patterns than BLSTM and BGRU. To our knowledge, no other framework is developed explicitly for evaluating splice detection models to decide the best possible model in an automated manner. So, the proposed framework and the blueprint would help selecting different deep learning models, such as CNN vs. BLSTM and BGRU, for splice site analysis or similar classification tasks and in different problems.

Modeling the combined effect of RNA-binding proteins and microRNAs in post-transcriptional regulation.

Recent studies show that RNA-binding proteins (RBPs) and microRNAs (miRNAs) function in coordination with each other to control post-transcriptional regulation (PTR). Despite this, the majority of research to date has focused on the regulatory effect of individual RBPs or miRNAs. Here, we mapped both RBP and miRNA binding sites on human 3'UTRs and utilized this collection to better understand PTR. We show that the transcripts that lack competition for HuR binding are destabilized more after HuR depletion. We also confirm this finding for PUM1(2) by measuring genome-wide expression changes following the knockdown of PUM1(2) in HEK293 cells. Next, to find potential cooperative interactions, we identified the pairs of factors whose sites co-localize more often than expected by random chance. Upon examining these results for PUM1(2), we found that transcripts where the sites of PUM1(2) and its interacting miRNA form a stem-loop are more stabilized upon PUM1(2) depletion. Finally, using dinucleotide frequency and counts of regulatory sites as features in a regression model, we achieved an AU-ROC of 0.86 in predicting mRNA half-life in BEAS-2B cells. Altogether, our results suggest that future studies of PTR must consider the combined effects of RBPs and miRNAs, as well as their interactions.

A novel SVM-ID3 hybrid feature selection method to build a disease model for melanoma using integrated genotyping and phenotype data from dbGaP.

The relations between Single Nucleotide Polymorphism (SNP) and complex diseases are likely to be non-linear and require analysis of the high dimensional data. Previous studies in the field mostly focus on genotyping and effects of various phenotypes are not considered. To fill this gap a hybrid feature selection model of support vector machine and decision tree has been designed. The designed method is tested on melanoma. We were able to select phenotypic features such as moles and dysplastic nevi, and SNPs those maps to specific genes such as CAMK1D. The performance results of the proposed hybrid model, on melanoma dataset are 79.07% of sensitivity and 0.81 of area under ROC curve.

Twenty-nail dystrophy in a mother and her 7-year-old daughter associated with balanced translocation 46, XX, t(6q13;10p13).

Twenty-nail dystrophy (TND) is an autosomal dominantly inherited idiopathic nail dystrophy characterized by excessive longitudinal striations and numerous superficial pits on nails with a typical 'sand papered' rough appearance. It is evident at birth and progresses slowly. It can also be associated with various diseases including lichen planus, alopecia areata, eczema, vitiligo and psoriasis. Peripheral blood chromosome analysis has not been performed in previously reported cases of TND. We report a mother and her 7-year-old daughter with TND. Both of them had a balanced translocation 46, XX, t(6q13;10p13). This may be a coincidental finding or may be related to the gene locus of TND.