Current Genomics - Volume 23, Issue 5, 2022

Genome sequences indicate a wide variety of characteristics, which include species and sub-species type, genotype, diseases, growth indicators, yield quality, etc. To analyze and study the characteristics of the genome sequences across different species, various deep learning models have been proposed by researchers, such as Convolutional Neural Networks (CNNs), Deep Belief Networks (DBNs), Multilayer Perceptrons (MLPs), etc., which vary in terms of evaluation performance, area of application and species that are processed. Due to a wide differentiation between the algorithmic implementations, it becomes difficult for research programmers to select the best possible genome processing model for their application. In order to facilitate this selection, the paper reviews a wide variety of such models and compares their performance in terms of accuracy, area of application, computational complexity, processing delay, precision and recall. Thus, in the present review, various deep learning and machine learning models have been presented that possess different accuracies for different applications. For multiple genomic data, Repeated Incremental Pruning to Produce Error Reduction with Support Vector Machine (Ripper SVM) outputs 99.7% of accuracy, and for cancer genomic data, it exhibits 99.27% of accuracy using the CNN Bayesian method. Whereas for Covid genome analysis, Bidirectional Long Short-Term Memory with CNN (BiLSTM CNN) exhibits the highest accuracy of 99.95%. A similar analysis of precision and recall of different models has been reviewed. Finally, this paper concludes with some interesting observations related to the genomic processing models and recommends applications for their efficient use.

Background: Circular RNAs (circRNAs) are transcribed by RNA polymerase II and are mostly generated by the back-splicing of exons in the protein-coding gene. Massive circRNAs are reported to be differentially expressed in different species, implicating their prospects as aging biomarkers or regulators in the aging progression. Methods: The possible role of circRNAs in aging and longevity was reviewed by the query of circRNAs from literature reports related to tissue, organ or cellular senescence, and individual longevity. Results: A number of circRNAs have been found to positively and negatively modulate aging and longevity through canonical aging pathways in the invertebrates Caenorhabditis elegans and Drosophila. Recent studies have also shown that circRNAs regulate age-related processes and pathologies such various mammalian tissues, as the brain, serum, heart, and muscle. Besides, three identified representative circRNAs (circSfl, circGRIA1, and circNF1-419) were elucidated to correlate with aging and longevity. Conclusion: This review outlined the current studies of circRNAs in aging and longevity, highlighting the role of circRNAs as a biomarker of aging and as a regulator of longevity.

In the decade since non-invasive prenatal testing (NIPT) was first implemented as a prenatal screening tool, it has gained recognition for its sensitivity and specificity in the detection of common aneuploidies. This review mainly focuses on the emerging role of NIPT in pregnancies following assisted reproductive technology (ART) in the light of current evidence and recommendations. It also deals with the challenges, shortcomings and interpretational difficulties related to NIPT in ART pregnancies, with particular emphasis on twin and vanishing twin pregnancies, which are widely regarded as the Achilles’ heel of most pre-natal screening platforms. Future directions for exploration towards improving the performance and extending the scope of NIPT are also addressed.

Preimplantation Genetic Testing (PGT) aims to reduce the chance of an affected pregnancy or improve success in an assisted reproduction cycle. Since the first established pregnancies in 1990, methodological approaches have greatly evolved, combined with significant advances in the embryological laboratory. The application of preimplantation testing has expanded, while the accuracy and reliability of monogenic and chromosomal analysis have improved. The procedure traditionally employs an invasive approach to assess the nucleic acid content of embryos. All biopsy procedures require high technical skill, and costly equipment, and may impact both the accuracy of genetic testing and embryo viability. To overcome these limitations, many researchers have focused on the analysis of cell-free DNA (cfDNA) at the preimplantation stage, sampled either from the blastocoel or embryo culture media, to determine the genetic status of the embryo non-invasively. Studies have assessed the origin of cfDNA and its application in non-invasive testing for monogenic disease and chromosomal aneuploidies. Herein, we discuss the state-of-the-art for modern non-invasive embryonic genetic material assessment in the context of PGT. The results are difficult to integrate due to numerous methodological differences between the studies, while further work is required to assess the suitability of cfDNA analysis for clinical application.

Background: One major challenge in binning Metagenomics data is the limited availability of reference datasets, as only 1% of the total microbial population is yet cultured. This has given rise to the efficacy of unsupervised methods for binning in the absence of any reference datasets. Objective: To develop a deep clustering-based binning approach for Metagenomics data and to evaluate results with suitable measures. Methods: In this study, a deep learning-based approach has been taken for binning the Metagenomics data. The results are validated on different datasets by considering features such as Tetra-nucleotide frequency (TNF), Hexa-nucleotide frequency (HNF) and GC-Content. Convolutional Autoencoder is used for feature extraction and for binning; the K-means clustering method is used. Results: In most cases, it has been found that evaluation parameters such as the Silhouette index and Rand index are more than 0.5 and 0.8, respectively, which indicates that the proposed approach is giving satisfactory results. The performance of the developed approach is compared with current methods and tools using benchmarked low complexity simulated and real metagenomic datasets. It is found better for unsupervised and at par with semi-supervised methods. Conclusion: An unsupervised advanced learning-based approach for binning has been proposed, and the developed method shows promising results for various datasets. This is a novel approach for solving the lack of reference data problem of binning in metagenomics.