Article
Biology
Roberto Serna-Blasco, Estela Sanchez-Herrero, Maria Berrocal Renedo, Silvia Calabuig-Farinas, Miguel Angel Molina-Vila, Mariano Provencio, Atocha Romero
Summary: Variant allele fraction and the median of absolute values of all pairwise differences impact the agreement between digital PCR and NGS calls. The new parameter R-score integrates these variables and can assist in optimizing NGS variant calling. There is a significant linear correlation between the PPA and the R-score, indicating that R-score can help in selecting reliable variants detected by NGS.
Article
Biology
Essam A. Rashed, Sachiko Kodera, Akimasa Hirata
Summary: Due to the variability of COVID-19 and various factors, it is challenging to predict its incidence using traditional mathematical models. In this study, machine learning was used to forecast the epidemic, taking into account the effectiveness of vaccination, infectivity of viral variants, and changes in public behavior. The results suggest that the effectiveness of vaccines and the infectivity of viral variants are important factors in predicting the incidence of COVID-19.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Pham Minh Thu Do, Thi Thanh Sang Nguyen
Summary: This paper proposes a novel semantic-enhanced Neural Collaborative Filtering (NCF) model for movie rating prediction and recommendation tasks. By building a semantic knowledge base and user behavior analytic model, combined with user preferences and recommendation model, the proposed model shows better recommendation performance in experiments.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Jesus Bobadilla, Fernando Ortega, Abraham Gutierrez, Angel Gonzalez-Prieto
Summary: The research introduces a method to incorporate stochasticity into deep learning models using variational autoencoders, aiming to improve the performance of recommender systems. By introducing variational techniques in the latent space, this approach can be applied as a plugin to current and future models, demonstrating superior performance in experiments.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Biochemical Research Methods
Yongzhuang Liu, Yalin Huang, Guohua Wang, Yadong Wang
Summary: Short read whole genome sequencing is widely used in human genetic studies and clinical practices for detecting structural variants, but accurate detection is challenging. This study introduces a novel deep learning-based approach, DeepSVFilter, for filtering structural variants in short read whole genome sequencing data.
BRIEFINGS IN BIOINFORMATICS
(2021)
Review
Biochemistry & Molecular Biology
Alfonso Monaco, Ester Pantaleo, Nicola Amoroso, Antonio Lacalamita, Claudio Lo Giudice, Adriano Fonzino, Bruno Fosso, Ernesto Picardi, Sabina Tangaro, Graziano Pesole, Roberto Bellotti
Summary: High throughput sequencing technologies have enabled the study of complex biological aspects at single nucleotide resolution, opening the big data era. The analysis of large volumes of heterogeneous omic data requires novel and efficient computational algorithms based on the paradigm of Artificial Intelligence. This review introduces and describes common machine learning methodologies, including deep learning, applied to various genomics tasks, highlighting the power of machine learning in handling big data and how these methods can be relevant in cases with large amounts of multimodal genomic data available.
COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL
(2021)
Article
Automation & Control Systems
Huafeng Liu, Liping Jing, Jingxuan Wen, Pengyu Xu, Jiaqi Wang, Jian Yu, Michael K. Ng
Summary: User preference modeling in recommendation system aims to improve customer experience, and deep generative models have been widely applied in this area. However, their capability in handling complex user preferences is limited.
JOURNAL OF MACHINE LEARNING RESEARCH
(2021)
Article
Biotechnology & Applied Microbiology
Guillaume P. Ramstein, Edward S. Buckler
Summary: This study accurately predicts nucleotide conservation across angiosperms using genomic annotations as a proxy for fitness effect of mutations. This approach can prioritize sites likely to impact fitness-related traits in crops and improve genomic prediction accuracy.
Article
Biotechnology & Applied Microbiology
Stephan Weissbach, Stanislav Sys, Charlotte Hewel, Hristo Todorov, Susann Schweiger, Jennifer Winter, Markus Pfenninger, Ali Torkamani, Doug Evans, Joachim Burger, Karin Everschor-Sitte, Helen Louise May-Simera, Susanne Gerber
Summary: The study demonstrates systematic heterogeneity in variant calls between different experimental and data analysis setups, and highlights the benefit of reprocessing genomic data with harmonized pipelines for improving concordance.
Article
Computer Science, Artificial Intelligence
Zongwei Zhou, Vatsal Sodha, Jiaxuan Pang, Michael B. Gotway, Jianming Liang
Summary: Transfer learning from natural image to medical image has been established as one of the most practical paradigms in deep learning for medical image analysis. To overcome the limitations of 3D imaging in prominent modalities like CT and MRI, a set of models called Models Genesis have been created to provide better performance in 3D medical imaging applications. The Models Genesis utilize self-supervised learning to automatically learn common anatomical representation, outperforming existing methods in both segmentation and classification tasks.
MEDICAL IMAGE ANALYSIS
(2021)
Article
Biochemistry & Molecular Biology
Mahdieh Labani, Amin Beheshti, Ahmadreza Argha, Hamid Alinejad-Rokny
Summary: Prostate cancer is highly prevalent and genomic alterations play a key role in its development and progression. In this study, an integrative analysis pipeline was used to identify 646 putative regulatory variants in prostate cancer, of which 30 significantly affected the expression of protein-coding genes. These variants could potentially impact 131 coding and non-coding genes, many of which are involved in disease-related pathways with targeted treatment options already available. Additionally, non-coding RNAs were identified as potential enhancer elements for certain protein-coding genes. Overall, this study provides a comprehensive map of genomic variants in prostate cancer and their potential contribution to the disease.
INTERNATIONAL JOURNAL OF MOLECULAR SCIENCES
(2023)
Article
Virology
Giulia Gatti, Martina Brandolini, Andrea Mancini, Francesca Taddei, Silvia Zannoli, Giorgio Dirani, Martina Manera, Valentina Arfilli, Agnese Denicolo, Anna Marzucco, Maria Sofia Montanari, Irene Zaghi, Massimiliano Guerra, Rita Tennina, Maria Michela Marino, Laura Grumiro, Monica Cricca, Vittorio Sambri
Summary: Since the first SARS-CoV-2 outbreak, mutations have changed the viral genome sequence, structure, and protein folding, leading to the onset of new variants. These alterations challenge both the clinical field and diagnostic demand due to detection failures and incomplete PCR results. Analyzing understudied genes such as N and investigating regions prone to mutation through WG-NGS can help identify new or reacquired mutations for designing robust primers.
Article
Biochemical Research Methods
Jiadong Lin, Songbo Wang, Peter A. Audano, Deyu Meng, Jacob Flores, Walter Kosters, Xiaofei Yang, Peng Jia, Tobias Marschall, Christine R. Beck, Kai Ye
Summary: SVision is a deep-learning-based tool that can automatically detect and characterize complex structural variants. It outperforms current methods in accurately identifying the internal structure of complex events and can detect both common and previously uncharacterized complex rearrangements.
Article
Biotechnology & Applied Microbiology
Rida Assaf, Fangfang Xia, Rick Stevens
Summary: We introduced a method called Shutter Island that utilizes a deep learning model (Inception V3) to detect genomic islands, showcasing better generalization than existing tools through image-based representation. The re-training of the model on a limited number of GI datasets and successful generalization suggests potential application of this approach to other data-limited fields.
Article
Computer Science, Artificial Intelligence
Fernando Moreno-Pino, Pablo M. Olmos, Antonio Artes-Rodriguez
Summary: In this paper, a forecasting architecture that combines deep autoregressive models with a Spectral Attention (SA) module is proposed. The architecture improves forecasting accuracy and produces explainable results in time series forecasting.
PATTERN RECOGNITION
(2023)
Article
Engineering, Biomedical
Cathleen Hagemann, Carmen Moreno Gonzalez, Ludovica Guetta, Giulia Tyzack, Ciro Chiappini, Andrea Legati, Rickie Patani, Andrea Serio
Summary: Stem cell-based experimental platforms in neuroscience can effectively mimic key aspects of human development and disease. However, conventional culture systems may not accurately represent the engineering constraints faced by cells in vivo, especially for neurons with long axons like spinal motor neurons. The establishment of a bioengineered platform to assemble arrays of human axons of various lengths has revealed a link between axon length and metabolism in human motor neurons, shedding light on a length-dependent mechanism that influences homeostatic processes within these cells. These findings have important implications for modeling neurodegenerative disorders in vitro and emphasize the importance of accurately modeling cell shape and biophysical constraints in experimental settings.
ADVANCED HEALTHCARE MATERIALS
(2022)
Article
Biochemistry & Molecular Biology
Edoardo Giacopuzzi, Niko Popitsch, Jenny C. Taylor
Summary: Non-coding variants play a significant role in common disease risks, and rare, high-impact non-coding variants are also accumulating. This article proposes a new framework for prioritizing non-coding regulatory variants by integrating information about regulatory regions, prediction scores, and HPO-based prioritization. The authors created a comprehensive collection of annotations for regulatory regions, calculated a variation constraint metric, compared non-coding impact prediction scores, and developed a VCF annotation tool. Evaluation results showed the effectiveness of the proposed framework.
NUCLEIC ACIDS RESEARCH
(2022)
Article
Psychiatry
Vincenzo Dattilo, Sheila Ulivi, Alessandra Minelli, Martina La Bianca, Edoardo Giacopuzzi, Marco Bortolomasi, Stefano Bignotti, Massimo Gennarelli, Paolo Gasparini, Maria Pina Concas
Summary: This study identified new possible genes associated with major depressive disorder (MDD) using a genome-wide association study (GWAS) approach. The findings confirmed the polygenic nature of MDD. Further research is needed to better understand the role of these genes in MDD.
WORLD JOURNAL OF BIOLOGICAL PSYCHIATRY
(2023)
Article
Cell Biology
Alessia Nasca, Andrea Legati, Megi Meneri, Melisa Emel Ermert, Chiara Frascarelli, Nadia Zanetti, Manuela Garbellini, Giacomo Pietro Comi, Alessia Catania, Costanza Lamperti, Dario Ronchi, Daniele Ghezzi
Summary: This study reports the identification of biallelic ENDOG variants in a patient with progressive external ophthalmoplegia, mitochondrial myopathy, and multiple mtDNA deletions in muscle. The absence of the ENDOG protein and the presence of multiple mtDNA deletions indicate the pathogenicity of the identified variants. The accumulation of low-level heteroplasmic mtDNA point mutations suggests a possible role of ENDOG in mtDNA replication or repair.
Article
Genetics & Heredity
L. Lenzini, M. Carecchio, E. Iori, A. Legati, E. Lamantea, A. Avogaro, N. Vitturi
Summary: This study reports a novel pathogenic variant in a patient with Leigh Syndrome, leading to reduced levels of mS34 protein and affecting OXPHOS complex. The patient exhibited delayed motor milestones, stepwise deterioration, and unusual long survival.
MOLECULAR GENETICS AND METABOLISM REPORTS
(2022)
Letter
Dermatology
Vincenzo Maione, Simone Soglia, Laura Miccio, Piergiacomo Calzavara-Pinton, Angela Napolitano, Valeria Cinquina, Marco Ritelli, Marina Colombi
JOURNAL DER DEUTSCHEN DERMATOLOGISCHEN GESELLSCHAFT
(2022)
Article
Biochemistry & Molecular Biology
Paola Fortugno, Rosanna Monetta, Valeria Cinquina, Chiara Rigon, Francesca Boaretto, Chiara De Luca, Nicoletta Zoppi, Luana Di Leandro, Emanuela De Domenico, Arianna Di Daniele, Rodolfo Ippoliti, Francesco Angelucci, Ernesto Di Cesare, Ruggero De Paulis, Leonardo Salviati, Marina Colombi, Francesco Brancati, Marco Ritelli
Summary: Pathogenic variants in TGFBR1 are a common cause of Loeys-Dietz syndrome (LDS) characterized by life-threatening cardiovascular diseases. In this study, two novel variants in TGFBR1 were identified in LDS patients, resulting in truncated TGFBR1 proteins. These variants escaped nonsense-mediated mRNA decay and exhibited enhanced TGF beta signaling. The findings emphasize the importance of functional studies for accurate clinical diagnosis.
EUROPEAN JOURNAL OF HUMAN GENETICS
(2023)
Article
Cell Biology
Marco Ritelli, Nicola Chiarelli, Valeria Cinquina, Nicoletta Zoppi, Valeria Bertini, Marina Venturini, Marina Colombi
Summary: This study characterized the cellular phenotype and gene expression profile of hypermobile Ehlers-Danlos syndrome (hEDS) and hypermobility spectrum disorders (HSD) dermal fibroblasts. The findings showed generalized extracellular matrix (ECM) disarray, myofibroblast differentiation, and dysregulated gene expression in both cell types. Based on these findings, a disease model was proposed in which an unbalanced ECM remodeling leads to functional impairment of different connective tissues in hEDS/HSD patients.
Article
Clinical Neurology
Alessia Nasca, Niccolo E. Mencacci, Federica Invernizzi, Michael Zech, Ignacio J. Keller Sarmiento, Andrea Legati, Chiara Frascarelli, Bernabe Bustos, Luigi M. Romito, Dimitri Krainc, Juliane Winkelmann, Miryam Carecchio, Nardo Nardocci, Giovanna Zorzi, Holger Prokisch, Steven J. Lubbe, Barbara Garavaglia, Daniele Ghezzi
Summary: Nasca et al. have discovered a new candidate gene for dystonia, ATP5F1B, which encodes a subunit of the mitochondrial ATP synthase. This gene is associated with early-onset isolated dystonia in two families with autosomal dominant inheritance and incomplete penetrance. Functional studies showed a dominant-negative effect of the identified ATP5F1B variants, leading to reduced activity of complex V and impaired mitochondrial function.
Review
Clinical Neurology
A. Legati, D. Ghezzi
Summary: This paper reviews the latest literature on the central role of mitochondrial dysfunction in Parkinson's disease pathophysiology, focusing on genetic defects and expression alterations affecting mitochondria-associated genes. Recent studies have shown that these alterations play a key role in the development of Parkinson's disease.
CURRENT NEUROLOGY AND NEUROSCIENCE REPORTS
(2023)
Article
Genetics & Heredity
Marco Ritelli, Nicola Chiarelli, Valeria Cinquina, Marika Vezzoli, Marina Venturini, Marina Colombi
Summary: The most common conditions with symptomatic joint hypermobility are hypermobile Ehlers-Danlos syndrome (hEDS) and hypermobility spectrum disorders (HSD). Diagnosing these overlapping connective tissue disorders remains challenging due to the lack of established causes and reliable diagnostic tests. The effectiveness of the 2017 diagnostic criteria in distinguishing between hEDS and HSD and the frequencies of extra-articular manifestations were evaluated retrospectively in this study.
AMERICAN JOURNAL OF MEDICAL GENETICS PART A
(2023)
Article
Genetics & Heredity
Chiara Frascarelli, Nadia Zanetti, Alessia Nasca, Rossella Izzo, Costanza Lamperti, Eleonora Lamantea, Andrea Legati, Daniele Ghezzi
Summary: Primary mitochondrial diseases are progressive genetic disorders characterized by mitochondrial dysfunction. They can be caused by mutations in nuclear genes or genetic defects in mitochondrial genome. Traditional sequencing methods are limited in detecting structural alterations in mtDNA, but new long-read NGS technologies show promise in this regard. This study presents optimized protocols using long-read Oxford Nanopore Technology for the detection of mtDNA structural alterations, which may become the method of choice for genetic studies on mtDNA.
FRONTIERS IN GENETICS
(2023)
Article
Neurosciences
Fabiana Colucci, Marcella Neri, Fernanda Fortunato, Alessandra Ferlini, Rosalba Carrozzo, Alessandra Torraco, Eleonora Lamantea, Andrea Legati, Ginevra Tecilla, Maura Pugliatti, Mariachiara Sensi
Summary: This study describes two patients with different clinical presentations of ataxia who carry the same biallelic mutation in the AFG3L2 gene.
Article
Rheumatology
M. Colman, M. Castori, L. Micale, M. Ritelli, M. Colombi, N. Ghali, F. Van Dijk, L. Marsili, A. Weeks, A. Vandersteen, A. Rideout, A. Legrand, M. Frank, T. Mirault, A. Ferraris, N. Di Giosaffatte, P. Grammatico, J. Grunert, C. Frank, S. Symoens, D. Syx, F. Malfait
Summary: This study reports on the phenotype and risk information of individuals carrying rare variants in the COL1A1 and COL3A1 genes. These rare variants are more common in individuals with classical and vascular Ehlers-Danlos syndrome, but their pathogenic effects are often difficult to predict.
CLINICAL AND EXPERIMENTAL RHEUMATOLOGY
(2022)
Article
Genetics & Heredity
Maribel Vazquez, Jack Chovanec, Jiwon Kim, Thomas DiMaggio, Joshua D. Milner, Clair A. Francomano, Christina A. Gurnett, Marco Ritelli, Marina Colombi, Jonathan J. Lyons
Summary: Hereditary alpha-tryptasemia (H alpha T) is a common genetic trait associated with joint hypermobility. Genotyping of individuals revealed significant associations between H alpha T and dysphagia and retained primary dentition.
HUMAN GENETICS AND GENOMICS ADVANCES
(2022)