Search

The Proteome-Wide Association Study (PWAS) is a protein-based genetic association approach designed to complement traditional variant-based methods like GWAS. PWAS operates in two stages: first, machine learning models predict the impact of genetic variants on protein-coding genes, generating effect scores. These scores are then aggregated into a gene-damaging score for each individual. This score is then used in case-control statistical tests to significantly link to specific phenotypes. PWAS Hub (v1.2) is a user-friendly platform that facilitates the exploration of gene-disease associations using clinical and genetic data from the UK Biobank (UKB), encompassing 500k individuals. PWAS Hub reports on 819 diseases and phenotypes determined by PheCode and ICD-10 clinical codes, each with a minimum of 400 affected individuals. PWAS-derived gene associations were reported for 72% of the tested phenotypes. The PWAS Hub also analyzes gene associations separately for males and females, considering sex-specific genetic effects, inheritance patterns (dominant and recessive), and gene pleiotropy. We illustrated the utility of the PWAS Hub for primary (essential) hypertension (I10), type 2 diabetes mellitus (E11), and specified haematuria (R31) that showed sex-dependent genetic signals. The PWAS Hub, available at pwas.huji.ac.il, is a valuable resource for studying genetic contributions to common diseases and sex-specific effects., (© The Author(s) 2024. Published by Oxford University Press on behalf of Nucleic Acids Research.)

PWAS (proteome-wide association study) is an innovative genetic association approach that complements widely used methods like GWAS (genome-wide association study). The PWAS approach involves consecutive phases. Initially, machine learning modeling and probabilistic considerations quantify the impact of genetic variants on protein-coding genes' biochemical functions. Secondly, for each individual, aggregating the variants per gene determines a gene-damaging score. Finally, standard statistical tests are activated in the case-control setting to yield statistically significant genes per phenotype. The PWAS Hub offers a user-friendly interface for an in-depth exploration of gene-disease associations from the UK Biobank (UKB). Results from PWAS cover 99 common diseases and conditions, each with over 10,000 diagnosed individuals per phenotype. Users can explore genes associated with these diseases, with separate analyses conducted for males and females. For each phenotype, the analyses account for sex-based genetic effects, inheritance modes (dominant and recessive), and the pleiotropic nature of associated genes. The PWAS Hub showcases its usefulness for asthma by navigating through proteomic-genetic analyses. Inspecting PWAS asthma-listed genes (a total of 27) provide insights into the underlying cellular and molecular mechanisms. Comparison of PWAS-statistically significant genes for common diseases to the Open Targets benchmark shows partial but significant overlap in gene associations for most phenotypes. Graphical tools facilitate comparing genetic effects between PWAS and coding GWAS results, aiding in understanding the sex-specific genetic impact on common diseases. This adaptable platform is attractive to clinicians, researchers, and individuals interested in delving into gene-disease associations and sex-specific genetic effects., (© 2024 Kelman et al.; Published by Cold Spring Harbor Laboratory Press.)

PARK7, also known as DJ-1, plays a critical role in protecting cells by functioning as a sensitive oxidation sensor and modulator of antioxidants. DJ-1 acts to maintain mitochondrial function and regulate transcription in response to different stressors. In this study, we showed that cell lines vary based on their antioxidation potential under basal conditions. The transcriptome of HEK293 cells was tested following knockdown (KD) of DJ-1 using siRNAs, which reduced the DJ-1 transcripts to only 12% of the original level. We compared the expression levels of 14k protein-coding transcripts and 4.2k non-coding RNAs relative to cells treated with non-specific siRNAs. Among the coding genes, approximately 200 upregulated differentially expressed genes (DEGs) signified a coordinated antiviral innate immune response. Most genes were associated with the regulation of type 1 interferons (IFN) and the induction of inflammatory cytokines. About a quarter of these genes were also induced in cells treated with non-specific siRNAs that were used as a negative control. Beyond the antiviral-like response, 114 genes were specific to the KD of DJ-1 with enrichment in RNA metabolism and mitochondrial functions. A smaller set of downregulated genes (58 genes) was associated with dysregulation in membrane structure, cell viability, and mitophagy. We propose that the KD DJ-1 perturbation diminishes the protective potency against oxidative stress. Thus, it renders the cells labile and responsive to the dsRNA signal by activating a large number of genes, many of which drive apoptosis, cell death, and inflammatory signatures. The KD of DJ-1 highlights its potency in regulating genes associated with antiviral responses, RNA metabolism, and mitochondrial functions, apparently through alteration in STAT activity and downstream signaling. Given that DJ-1 also acts as an oncogene in metastatic cancers, targeting DJ-1 could be a promising therapeutic strategy where manipulation of the DJ-1 level may reduce cancer cell viability and enhance the efficacy of cancer treatments.

Background/objectives: The effects of early life exposures on offspring life-course health are well established. This study assessed whether adding early socio-demographic and perinatal variables to a model based on polygenic risk score (PRS) improves prediction of obesity risk., Methods: We used the Jerusalem Perinatal study (JPS) with data at birth and body mass index (BMI) and waist circumference (WC) measured at age 32. The PRS was constructed using over 2.1M common SNPs identified in genome-wide association study (GWAS) for BMI. Linear and logistic models were applied in a stepwise approach. We first examined the associations between genetic variables and obesity-related phenotypes (e.g., BMI and WC). Secondly, socio-demographic variables were added and finally perinatal exposures, such as maternal pre-pregnancy BMI (mppBMI) and gestational weight gain (GWG) were added to the model. Improvement in prediction of each step was assessed using measures of model discrimination (area under the curve, AUC), net reclassification improvement (NRI) and integrated discrimination improvement (IDI)., Results: One standard deviation (SD) change in PRS was associated with a significant increase in BMI (β = 1.40) and WC (β = 2.45). These associations were slightly attenuated (13.7-14.2%) with the addition of early life exposures to the model. Also, higher mppBMI was associated with increased offspring BMI (β = 0.39) and WC (β = 0.79) (p < 0.001). For obesity (BMI ≥ 30) prediction, the addition of early socio-demographic and perinatal exposures to the PRS model significantly increased AUC from 0.69 to 0.73. At an obesity risk threshold of 15%, the addition of early socio-demographic and perinatal exposures to the PRS model provided a significant improvement in reclassification of obesity (NRI, 0.147; 95% CI 0.068-0.225)., Conclusions: Inclusion of early life exposures, such as mppBMI and maternal smoking, to a model based on PRS improves obesity risk prediction in an Israeli population-sample., (© 2024. The Author(s).)

Hypothyroidism is a common endocrine disorder whose prevalence increases with age. The disease manifests itself when the thyroid gland fails to produce sufficient thyroid hormones. The disorder includes cases of congenital hypothyroidism (CH), but most cases exhibit hormonal feedback dysregulation and destruction of the thyroid gland by autoantibodies. In this study, we sought to identify causal genes for hypothyroidism in large populations. The study used the UK-Biobank (UKB) database, reporting on 13,687 cases of European ancestry. We used GWAS compilation from Open Targets (OT) and tuned protocols focusing on genes and coding regions, along with complementary association methods of PWAS (proteome-based) and TWAS (transcriptome-based). Comparing summary statistics from numerous GWAS revealed a limited number of variants associated with thyroid development. The proteome-wide association study method identified 77 statistically significant genes, half of which are located within the Chr6-MHC locus and are enriched with autoimmunity-related genes. While coding GWAS and PWAS highlighted the centrality of immune-related genes, OT and transcriptome-wide association study mostly identified genes involved in thyroid developmental programs. We used independent populations from Finland (FinnGen) and the Taiwan cohort to validate the PWAS results. The higher prevalence in females relative to males is substantiated as the polygenic risk score prediction of hypothyroidism relied mostly from the female group genetics. Comparing results from OT, TWAS, and PWAS revealed the complementary facets of hypothyroidism's etiology. This study underscores the significance of synthesizing gene-phenotype association methods for this common, intricate disease. We propose that the integration of established association methods enhances interpretability and clinical utility., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Zucker, Kovalerchik, Stern, Kaufman and Linial.)

Background: Distinguishing diseases into distinct subtypes is crucial for study and effective treatment strategies. The Open Targets Platform (OT) integrates biomedical, genetic, and biochemical datasets to empower disease ontologies, classifications, and potential gene targets. Nevertheless, many disease annotations are incomplete, requiring laborious expert medical input. This challenge is especially pronounced for rare and orphan diseases, where resources are scarce., Methods: We present a machine learning approach to identifying diseases with potential subtypes, using the approximately 23,000 diseases documented in OT. We derive novel features for predicting diseases with subtypes using direct evidence. Machine learning models were applied to analyze feature importance and evaluate predictive performance for discovering both known and novel disease subtypes., Results: Our model achieves a high (89.4%) ROC AUC (Area Under the Receiver Operating Characteristic Curve) in identifying known disease subtypes. We integrated pre-trained deep-learning language models and showed their benefits. Moreover, we identify 515 disease candidates predicted to possess previously unannotated subtypes., Conclusions: Our models can partition diseases into distinct subtypes. This methodology enables a robust, scalable approach for improving knowledge-based annotations and a comprehensive assessment of disease ontology tiers. Our candidates are attractive targets for further study and personalized medicine, potentially aiding in the unveiling of new therapeutic indications for sought-after targets., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024. Published by Elsevier Inc.)

Breast cancer (BC) is the most common malignancy affecting Western women today. It is estimated that as many as 10% of BC cases can be attributed to germline variants. However, the genetic basis of the majority of familial BC cases has yet to be identified. Discovering predisposing genes contributing to familial BC is challenging due to their presumed rarity, low penetrance, and complex biological mechanisms. Here, we focused on an analysis of rare missense variants in a cohort of 12 families of Middle Eastern origins characterized by a high incidence of BC cases. We devised a novel, high-throughput, variant analysis pipeline adapted for family studies, which aims to analyze variants at the protein level by employing state-of-the-art machine learning models and three-dimensional protein structural analysis. Using our pipeline, we analyzed 1218 rare missense variants that are shared between affected family members and classified 80 genes as candidate pathogenic. Among these genes, we found significant functional enrichment in peroxisomal and mitochondrial biological pathways which segregated across seven families in the study and covered diverse ethnic groups. We present multiple evidence that peroxisomal and mitochondrial pathways play an important, yet underappreciated, role in both germline BC predisposition and BC survival., (© The Author(s) 2024. Published by Oxford University Press.)

Many advances in biomedicine can be attributed to identifying unusual proteins and genes. Many of these proteins' unique properties were discovered by manual inspection, which is becoming infeasible at the scale of modern protein datasets. Here, we propose to tackle this challenge using anomaly detection methods that automatically identify unexpected properties. We adopt a state-of-the-art anomaly detection paradigm from computer vision, to highlight unusual proteins. We generate meaningful representations without labeled inputs, using pretrained deep neural network models. We apply these protein language models (pLM) to detect anomalies in function, phylogenetic families, and segmentation tasks. We compute protein anomaly scores to highlight human prion-like proteins, distinguish viral proteins from their host proteome, and mark non-classical ion/metal binding proteins and enzymes. Other tasks concern segmentation of protein sequences into folded and unstructured regions. We provide candidates for rare functionality (e.g. prion proteins). Additionally, we show the anomaly score is useful in 3D folding-related segmentation. Our novel method shows improved performance over strong baselines and has objectively high performance across a variety of tasks. We conclude that the combination of pLM and anomaly detection techniques is a valid method for discovering a range of global and local protein characteristics., (© The Author(s) 2024. Published by Oxford University Press on behalf of NAR Genomics and Bioinformatics.)

Treatment of aging rats for 6 months with ladostigil (1 mg/kg/day) prevented a decline in recognition and spatial memory and suppressed the overexpression of gene-encoding pro-inflammatory cytokines, TNFα, IL1β, and IL6 in the brain and microglial cultures. Primary cultures of mouse microglia stimulated by lipopolysaccharides (LPS, 0.75 µg/mL) and benzoyl ATPs (BzATP) were used to determine the concentration of ladostigil that reduces the secretion of these cytokine proteins. Ladostigil (1 × 10 -11 M), a concentration compatible with the blood of aging rats in, prevented memory decline and reduced secretion of IL1β and IL6 by ≈50%. RNA sequencing analysis showed that BzATP/LPS upregulated 25 genes, including early-growth response protein 1, (Egr1) which increased in the brain of subjects with neurodegenerative diseases. Ladostigil significantly decreased Egr1 gene expression and levels of the protein in the nucleus and increased TNF alpha-induced protein 3 (TNFaIP3), which suppresses cytokine release, in the microglial cytoplasm. Restoration of the aberrant signaling of these proteins in ATP/LPS-activated microglia in vivo might explain the prevention by ladostigil of the morphological and inflammatory changes in the brain of aging rats.

Scientific research trends and interests evolve over time. The ability to identify and forecast these trends is vital for educational institutions, practitioners, investors, and funding organizations. In this study, we predict future trends in scientific publications using heterogeneous sources, including historical publication time series from PubMed, research and review articles, pre-trained language models, and patents. We demonstrate that scientific topic popularity levels and changes (trends) can be predicted five years in advance across 40 years and 125 diverse topics, including life-science concepts, biomedical, anatomy, and other science, technology, and engineering topics. Preceding publications and future patents are leading indicators for emerging scientific topics. We find the ratio of reviews to original research articles informative for identifying increasing or declining topics, with declining topics having an excess of reviews. We find that language models provide improved insights and predictions into temporal dynamics. In temporal validation, our models substantially outperform the historical baseline. Our findings suggest that similar dynamics apply across other scientific and engineering research topics. We present SciTrends, a user-friendly webtool for predicting future publication trends for any topic covered in PubMed., Competing Interests: The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:Michal Linial reports financial support was provided by 10.13039/501100003483Hebrew University of Jerusalem. Michal Linial reports a relationship with Hebrew University of Jerusalem that includes: employment. None., (© 2023 Published by Elsevier Ltd.)

We assessed whether adding early life exposures to a model based on polygenic risk score (PRS) improves prediction of obesity risk. We used a birth cohort with data at birth and BMI and waist circumference (WC) measured at age 32. The PRS was composed of SNPs identified in GWAS for BMI. Linear and logistic models were used to explore associations with obesity-related phenotypes. Improvement in prediction was assessed using measures of model discrimination (AUC), and net reclassification improvement (NRI). One SD change in PRS was associated with a significant increase in BMI and WC. These associations were slightly attenuated (13.7%-14.2%) with the addition of early life exposures to the model. Also, higher maternal pre-pregnancy BMI was associated with increase in offspring BMI and WC (p<0.001). For prediction obesity (BMI ≥ 30), the addition of early life exposures to the PRS model significantly increase the AUC from 0.69 to 0.73. At an obesity risk threshold of 15%, the addition of early life exposures to the PRS model provided a significant improvement in reclassification of obesity (NRI, 0.147; 95% CI 0.068-0.225). We conclude that inclusion of early life exposures to a model based on PRS improves obesity risk prediction in an Israeli population-sample., Competing Interests: Disclosure: The authors declared no conflict of interest.

Hypertension is a polygenic disease that affects over 1.2 billion adults aged 30-79 worldwide. It is a major risk factor for renal, cerebrovascular, and cardiovascular diseases. The heritability of hypertension is estimated to be high; nevertheless, our understanding of its underlying mechanisms remains scarce and incomplete. This study covered the entries from European ancestry from the UK-Biobank (UKB), with 74,090 cases diagnosed with essential (primary) hypertension and 200,734 controls. We compared the findings from large-scale genome-wide association studies (GWAS) to the gene-based method of proteome-wide association studies (PWAS). We focused on 70 statistically significant associated genes, most of which failed to reach significance in variant-based GWAS. A total of 30% of the PWAS-associated genes were validated against independent cohorts, including the Finnish Biobank. Furthermore, gene-based analyses that were performed on both sexes revealed sex-dependent genetics with a stronger genetic component associated with females. Analysis of systolic and diastolic blood pressure measurements confirms a strong genetic effect associated with females. We demonstrated that gene-based approaches provide insight into the underlying biology of hypertension. Specifically, the expression profiles of the identified genes exposed the enrichment of endothelial cells from multiple organs. Furthermore, females' top-ranked significant genes are involved in cellular immunity. We conclude that studying hypertension and blood pressure via gene-based association methods improves interpretability and exposes sex-dependent genetic effects, which enhances clinical utility., (© 2023. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.)

The primary role of microglia is to maintain homeostasis by effectively responding to various disturbances. Activation of transcriptional programs determines the microglia's response to external stimuli. In this study, we stimulated murine neonatal microglial cells with benzoyl ATP (bzATP) and lipopolysaccharide (LPS), and monitored their ability to release pro-inflammatory cytokines. When cells are exposed to bzATP, a purinergic receptor agonist, a short-lived wave of transcriptional changes, occurs. However, only combining bzATP and LPS led to a sustainable and robust response. The transcriptional profile is dominated by induced cytokines (e.g., IL-1α and IL-1β), chemokines, and their membrane receptors. Several abundant long noncoding RNAs (lncRNAs) are induced by bzATP/LPS, including Ptgs2os2, Bc1, and Morrbid, that function in inflammation and cytokine production. Analyzing the observed changes through TNF (Tumor necrosis factor) and NF-κB (nuclear factor kappa light chain enhancer of activated B cells) pathways confirmed that neonatal glial cells exhibit a distinctive expression program in which inflammatory-related genes are upregulated by orders of magnitude. The observed capacity of the microglial culture to activate a robust inflammatory response is useful for studying neurons under stress, brain injury, and aging. We propose the use of a primary neonatal microglia culture as a responsive in vitro model for testing drugs that may interact with inflammatory signaling and the lncRNA regulatory network.

Mature microRNAs (miRNAs) are single-stranded non-coding RNA (ncRNA) molecules that act in post-transcriptional regulation in animals and plants. A mature miRNA is the end product of consecutive, highly regulated processing steps of the primary miRNA transcript. Following base-paring of the mature miRNA with its mRNA target, translation is inhibited, and the targeted mRNA is degraded. There are hundreds of miRNAs in each cell that work together to regulate cellular key processes, including development, differentiation, cell cycle, apoptosis, inflammation, viral infection, and more. In this review, we present an overlooked layer of cellular regulation that addresses cell dynamics affecting miRNA accessibility. We discuss the regulation of miRNA local storage and translocation among cell compartments. The local amounts of the miRNAs and their targets dictate their actual availability, which determines the ability to fine-tune cell responses to abrupt or chronic changes. We emphasize that changes in miRNA storage and compactization occur under induced stress and changing conditions. Furthermore, we demonstrate shared principles on cell physiology, governed by miRNA under oxidative stress, tumorigenesis, viral infection, or synaptic plasticity. The evidence presented in this review article highlights the importance of spatial and temporal miRNA regulation for cell physiology. We argue that limiting the research to mature miRNAs within the cytosol undermines our understanding of the efficacy of miRNAs to regulate cell fate under stress conditions.

Neurodegenerative disorders, brain injury, and the decline in cognitive function with aging are accompanied by a reduced capacity of cells in the brain to cope with oxidative stress and inflammation. In this study, we focused on the response to oxidative stress in SH-SY5Y, a human neuroblastoma cell line. We monitored the viability of the cells in the presence of oxidative stress. Such stress was induced by hydrogen peroxide or by Sin1 (3-morpholinosydnonimine) that generates reactive oxygen and nitrogen species (ROS and RNS). Both stressors caused significant cell death. Our results from the RNA-seq experiments show that SH-SY5Y cells treated with Sin1 for 24 h resulted in 94 differently expressed long non-coding RNAs (lncRNAs), including many abundant ones. Among the abundant lncRNAs that were upregulated by exposing the cells to Sin1 were those implicated in redox homeostasis, energy metabolism, and neurodegenerative diseases (e.g., MALAT1, MIAT, GABPB1-AS1, NEAT1, MIAT, GABPB1-AS1, and HAND2-AS1). Another group of abundant lncRNAs that were significantly altered under oxidative stress included cancer-related SNHG family members. We tested the impact of ladostigil, a bifunctional reagent with antioxidant and anti-inflammatory properties, on the lncRNA expression levels. Ladostigil was previously shown to enhance learning and memory in the brains of elderly rats. In SH-SY5Y cells, several lncRNAs involved in transcription regulation and the chromatin structure were significantly induced by ladostigil. We anticipate that these poorly studied lncRNAs may act as enhancers (eRNA), regulating transcription and splicing, and in competition for miRNA binding (ceRNA). We found that the induction of abundant lncRNAs, such as MALAT1, NEAT-1, MIAT, and SHNG12, by the Sin1 oxidative stress paradigm specifies only the undifferentiated cell state. We conclude that a global alteration in the lncRNA profiles upon stress in SH-SY5Y may shift cell homeostasis and is an attractive in vitro system to characterize drugs that impact the redox state of the cells and their viability.

Post-transcriptional regulation in multicellular organisms is mediated by microRNAs. However, the principles that determine if a gene is regulated by miRNAs are poorly understood. Previous works focused mostly on miRNA seed matches and other features of the 3'-UTR of transcripts. These common approaches rely on knowledge of the miRNA families, and computational approaches still yield poor, inconsistent results, with many false positives. In this work, we present a different paradigm for predicting miRNA-regulated genes based on the encoded proteins. In a novel, automated machine learning framework, we use sequence as well as diverse functional annotations to train models on multiple organisms using experimentally validated data. We present insights from tens of millions of features extracted and ranked from different modalities. We show high predictive performance per organism and in generalization across species. We provide a list of novel predictions including Danio rerio (zebrafish) and Arabidopsis thaliana (mouse-ear cress). We compare genomic models and observe that our protein model outperforms, whereas a unified model improves on both. While most membranous and disease related proteins are regulated by miRNAs, the G-protein coupled receptor (GPCR) family is an exception, being mostly unregulated by miRNAs. We further show that the evolutionary conservation among paralogs does not imply any coherence in miRNA regulation. We conclude that duplicated paralogous genes that often changed their function, also diverse in their tendency to be miRNA regulated. We conclude that protein function is informative across species in predicting post-transcriptional miRNA regulation in living cells., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Ofer and Linial.)

Endometriosis is a condition characterized by implants of endometrial tissues into extrauterine sites, mostly within the pelvic peritoneum. The prevalence of endometriosis is under-diagnosed and is estimated to account for 5-10% of all women of reproductive age. The goal of this study was to develop a model for endometriosis based on the UK-biobank (UKB) and re-assess the contribution of known risk factors to endometriosis. We partitioned the data into those diagnosed with endometriosis (5924; ICD-10: N80) and a control group (142,723). We included over 1000 variables from the UKB covering personal information about female health, lifestyle, self-reported data, genetic variants, and medical history prior to endometriosis diagnosis. We applied machine learning algorithms to train an endometriosis prediction model. The optimal prediction was achieved with the gradient boosting algorithms of CatBoost for the data-combined model with an area under the ROC curve (ROC-AUC) of 0.81. The same results were obtained for women from a mixed ethnicity population of the UKB (7112; ICD-10: N80). We discovered that, prior to being diagnosed with endometriosis, affected women had significantly more ICD-10 diagnoses than the average unaffected woman. We used SHAP, an explainable AI tool, to estimate the marginal impact of a feature, given all other features. The informative features ranked by SHAP values included irritable bowel syndrome (IBS) and the length of the menstrual cycle. We conclude that the rich population-based retrospective data from the UKB are valuable for developing unified machine learning endometriosis models despite the limitations of missing data, noisy medical input, and participant age. The informative features of the model may improve clinical utility for endometriosis diagnosis.

Genetic studies of human traits have revolutionized our understanding of the variation between individuals, and yet, the genetics of most traits is still poorly understood. In this review, we highlight the major open problems that need to be solved, and by discussing these challenges provide a primer to the field. We cover general issues such as population structure, epistasis and gene-environment interactions, data-related issues such as ancestry diversity and rare genetic variants, and specific challenges related to heritability estimates, genetic association studies, and polygenic risk scores. We emphasize the interconnectedness of these problems and suggest promising avenues to address them., (© 2022. The Author(s).)

Summary: Self-supervised deep language modeling has shown unprecedented success across natural language tasks, and has recently been repurposed to biological sequences. However, existing models and pretraining methods are designed and optimized for text analysis. We introduce ProteinBERT, a deep language model specifically designed for proteins. Our pretraining scheme combines language modeling with a novel task of Gene Ontology (GO) annotation prediction. We introduce novel architectural elements that make the model highly efficient and flexible to long sequences. The architecture of ProteinBERT consists of both local and global representations, allowing end-to-end processing of these types of inputs and outputs. ProteinBERT obtains near state-of-the-art performance, and sometimes exceeds it, on multiple benchmarks covering diverse protein properties (including protein structure, post-translational modifications and biophysical attributes), despite using a far smaller and faster model than competing deep-learning methods. Overall, ProteinBERT provides an efficient framework for rapidly training protein predictors, even with limited labeled data., Availability and Implementation: Code and pretrained model weights are available at https://github.com/nadavbra/protein&#95;bert., Supplementary Information: Supplementary data are available at Bioinformatics online., (© The Author(s) 2022. Published by Oxford University Press.)

MicroRNAs (miRNAs) provide a fundamental layer of regulation in cells. miRNAs act posttranscriptionally through complementary base-pairing with the 3'-UTR of a target mRNA, leading to mRNA degradation and translation arrest. The likelihood of forming a valid miRNA-target duplex within cells was computationally predicted and experimentally monitored. In human cells, the miRNA profiles determine their identity and physiology. Therefore, alterations in the composition of miRNAs signify many cancer types and chronic diseases. In this chapter, we introduce online functional tools and resources to facilitate miRNA research. We start by introducing currently available miRNA catalogs and miRNA-gateway portals for navigating among different miRNA-centric online resources. We then sketch several realistic challenges that may occur while investigating miRNA regulation in living cells. As a showcase, we demonstrate the utility of miRNAs and mRNAs expression databases that cover diverse human cells and tissues, including resources that report on genetic alterations affecting miRNA expression levels and alteration in binding capacity. Introducing tools linking miRNAs with transcription factor (TF) networks reveals miRNA regulation complexity within living cells. Finally, we concentrate on online resources that analyze miRNAs in human diseases and specifically in cancer. Altogether, we introduce contemporary, selected resources and online tools for studying miRNA regulation in cells and tissues and their utility in health and disease., (© 2022. The Author(s), under exclusive license to Springer Nature Switzerland AG.)

A hallmark of cancer evolution is that the tumor may change its cell identity and improve its survival and fitness. Drastic change in microRNA (miRNA) composition and quantities accompany such dynamic processes. Cancer samples are composed of cells' mixtures of varying stages of cancerous progress. Therefore, cell-specific molecular profiling represents cellular averaging. In this study, we consider the degree to which altering miRNAs composition shifts cell behavior. We used COMICS, an iterative framework that simulates the stochastic events of miRNA-mRNA pairing, using a probabilistic approach. COMICS simulates the likelihood that cells change their transcriptome following many iterations (100 k). Results of COMICS from the human cell line (HeLa) confirmed that most genes are resistant to miRNA regulation. However, COMICS results suggest that the composition of the abundant miRNAs dictates the nature of the cells (across three cell lines) regardless of its actual mRNA steady-state. In silico perturbations of cell lines (i.e., by overexpressing miRNAs) allowed to classify genes according to their sensitivity and resilience to any combination of miRNA perturbations. Our results expose an overlooked quantitative dimension for a set of genes and miRNA regulation in living cells. The immediate implication is that even relatively modest overexpression of specific miRNAs may shift cell identity and impact cancer evolution., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Mahlab-Aviv, Linial and Linial.)

Background: Hepatitis E (HEV) is an emerging cause of viral hepatitis worldwide. Swine carrying hepatitis E genotype 3 (HEV-3) are responsible for the majority of chronic viral hepatitis cases in developed countries. Recently, genotype 7 (HEV-7), isolated from a dromedary camel in the United Arab Emirates, was also associated with chronic viral hepatitis in a transplant recipient. In Israel, chronic HEV infection has not yet been reported, although HEV seroprevalence in humans is ~10%. Camels and swine are >65% seropositive. Here we report on the isolation and characterization of HEV from local camels and swine., Methods: Sera from camels ( n = 142), feces from swine ( n = 18) and blood from patients suspected of hepatitis E ( n = 101) were collected during 2017-2020 and used to detect and characterize HEV sequences., Results: HEV-3 isolated from local swine and the camel-derived HEV-7 sequence were highly similar to HEV-3f and HEV-7 sequences (88.2% and 86.4%, respectively) related to viral hepatitis. The deduced amino acid sequences of both isolates were also highly conserved (>98%). Two patients were HEV-RNA positive; acute HEV-1 infection could be confirmed in one of them., Discussion: The absence of any reported HEV-3 and HEV-7 infection in humans remains puzzling, especially considering the reported seroprevalence rates, the similarity between HEV sequences related to chronic hepatitis and the HEV genotypes identified in swine and camels in Israel.

A hallmark of the aging brain is the robust inflammation mediated by microglial activation. Pathophysiology of common neurodegenerative diseases involves oxidative stress and neuroinflammation. Chronic treatment of aging rats by ladostigil, a compound with antioxidant and anti-inflammatory function, prevented microglial activation and learning deficits. In this study, we further investigate the effect of ladostigil on undifferentiated SH-SY5Y cells. We show that SH-SY5Y cells exposed to acute (by H 2 O 2 ) or chronic oxidative stress (by Sin1, 3-morpholinosydnonimine) induced apoptotic cell death. However, in the presence of ladostigil, the decline in cell viability and the increase of oxidative levels were partially reversed. RNA-seq analysis showed that prolonged oxidation by Sin1 resulted in a simultaneous reduction of the expression level of endoplasmic reticulum (ER) genes that participate in proteostasis. By comparing the differential gene expression profile of Sin1 treated cells to cells incubated with ladostigil before being exposed to Sin1, we observed an over-expression of Clk1 (Cdc2-like kinase 1) which was implicated in psychophysiological stress in mice and Alzheimer's disease. Ladostigil also suppressed the expression of Ccpg1 (Cell cycle progression 1) and Synj1 (Synaptojanin 1) that are involved in ER-autophagy and endocytic pathways. We postulate that ladostigil alleviated cell damage induced by oxidation. Therefore, under conditions of chronic stress that are observed in the aging brain, ladostigil may block oxidative stress processes and consequently reduce neurotoxicity.

Human genetic variation in coding regions is fundamental to the study of protein structure and function. Most methods for interpreting missense variants consider substitution measures derived from homologous proteins across different species. In this study, we introduce human-specific amino acid (AA) substitution matrices that are based on genetic variations in the modern human population. We analyzed the frequencies of >4.8M single nucleotide variants (SNVs) at codon and AA resolution and compiled human-centric substitution matrices that are fundamentally different from classic cross-species matrices (e.g. BLOSUM, PAM). Our matrices are asymmetric, with some AA replacements showing significant directional preference. Moreover, these AA matrices are only partly predicted by nucleotide substitution rates. We further test the utility of our matrices in exposing functional signals of experimentally-validated protein annotations. A significant reduction in AA transition frequencies was observed across nine post-translational modification (PTM) types and four ion-binding sites. Our results propose a purifying selection signal in the human proteome across a diverse set of functional protein annotations and provide an empirical baseline for interpreting human genetic variation in coding regions., (© The Author(s) 2021. Published by Oxford University Press on behalf of NAR Genomics and Bioinformatics.)

Atomic structures of several proteins from the coronavirus family are still partial or unavailable. A possible reason for this gap is the instability of these proteins outside of the cellular context, thereby prompting the use of in-cell approaches. In situ cross-linking and mass spectrometry (in situ CLMS) can provide information on the structures of such proteins as they occur in the intact cell. Here, we applied targeted in situ CLMS to structurally probe Nsp1, Nsp2, and nucleocapsid (N) proteins from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and obtained cross-link sets with an average density of one cross-link per 20 residues. We then employed integrative modeling that computationally combined the cross-linking data with domain structures to determine full-length atomic models. For the Nsp2, the cross-links report on a complex topology with long-range interactions. Integrative modeling with structural prediction of individual domains by the AlphaFold2 system allowed us to generate a single consistent all-atom model of the full-length Nsp2. The model reveals three putative metal binding sites and suggests a role for Nsp2 in zinc regulation within the replication-transcription complex. For the N protein, we identified multiple intra- and interdomain cross-links. Our integrative model of the N dimer demonstrates that it can accommodate three single RNA strands simultaneously, both stereochemically and electrostatically. For the Nsp1, cross-links with the 40S ribosome were highly consistent with recent cryogenic electron microscopy structures. These results highlight the importance of cellular context for the structural probing of recalcitrant proteins and demonstrate the effectiveness of targeted in situ CLMS and integrative modeling., Competing Interests: The authors declare no competing interest., (Copyright   2021 the Author(s). Published by PNAS.)

Emerging SARS-CoV-2 variants may threaten global vaccination efforts and the awaited reduction in outbreak burden. In this study, we report a novel variant carrying the L452R mutation that emerged from a local B.1.362 lineage, B.1.362+L452R. The L452R mutation is associated with the Delta and Epsilon variants and was shown to cause increased infection and reduction in neutralization in pseudoviruses. Indeed, the B.1.362+L452R variant demonstrated a X4-fold reduction in neutralization capacity of sera from BNT162b2-vaccinated individuals compared to a wild-type strain. The variant infected 270 individuals in Israel between December 2020 and March 2021, until diminishing due to the gain in dominance of the Alpha variant in February 2021. This study demonstrates an independent, local emergence of a variant carrying a critical mutation, L452R, which may have the potential of becoming a variant of concern and emphasizes the importance of routine surveillance and detection of novel variants among efforts undertaken to prevent further disease spread.

During the past decade, whole-genome sequencing of tumor biopsies and individuals with congenital disorders highlighted the phenomenon of chromoanagenesis, a single chaotic event of chromosomal rearrangement. Chromoanagenesis was shown to be frequent in many types of cancers, to occur in early stages of cancer development, and significantly impact the tumor's nature. However, an in-depth, cancer-type dependent analysis has been somewhat incomplete due to the shortage in whole genome sequencing of cancerous samples. In this study, we extracted data from The Pan-Cancer Analysis of Whole Genome (PCAWG) and The Cancer Genome Atlas (TCGA) to construct and test a machine learning algorithm that can detect chromoanagenesis with high accuracy (86%). The algorithm was applied to ~10,000 unlabeled TCGA cancer patients. We utilize the chromoanagenesis assignment results, to analyze cancer-type specific chromoanagenesis characteristics in 20 TCGA cancer types. Our results unveil prominent genes affected in either chromoanagenesis or non-chromoanagenesis tumorigenesis. The analysis reveals a mutual exclusivity relationship between the genes impaired in chromoanagenesis versus non-chromoanagenesis cases. We offer the discovered characteristics as possible targets for cancer diagnostic and therapeutic purposes.

The characterization of germline genetic variation affecting cancer risk, known as cancer predisposition, is fundamental to preventive and personalized medicine. Studies of genetic cancer predisposition typically identify significant genomic regions based on family-based cohorts or genome-wide association studies (GWAS). However, the results of such studies rarely provide biological insight or functional interpretation. In this study, we conducted a comprehensive analysis of cancer predisposition in the UK Biobank cohort using a new gene-based method for detecting protein-coding genes that are functionally interpretable. Specifically, we conducted proteome-wide association studies (PWAS) to identify genetic associations mediated by alterations to protein function. With PWAS, we identified 110 significant gene-cancer associations in 70 unique genomic regions across nine cancer types and pan-cancer. In 48 of the 110 PWAS associations (44%), estimated gene damage is associated with reduced rather than elevated cancer risk, suggesting a protective effect. Together with standard GWAS, we implicated 145 unique genomic loci with cancer risk. While most of these genomic regions are supported by external evidence, our results also highlight many novel loci. Based on the capacity of PWAS to detect non-additive genetic effects, we found that 46% of the PWAS-significant cancer regions exhibited exclusive recessive inheritance. These results highlight the importance of recessive genetic effects, without relying on familial studies. Finally, we show that many of the detected genes exert substantial cancer risk in the studied cohort determined by a quantitative functional description, suggesting their relevance for diagnosis and genetic consulting., (© 2021. The Author(s).)

One of the major challenges in the post-genomic era is elucidating the genetic basis of human diseases. In recent years, studies have shown that polygenic risk scores ( PRS ), based on aggregated information from millions of variants across the human genome, can estimate individual risk for common diseases. In practice, the current medical practice still predominantly relies on physiological and clinical indicators to assess personal disease risk. For example, caregivers mark individuals with high body mass index (BMI) as having an increased risk to develop type 2 diabetes (T2D). An important question is whether combining PRS with clinical metrics can increase the power of disease prediction in particular from early life. In this work we examined this question, focusing on T2D. We present here a sex-specific integrated approach that combines PRS with additional measurements and age to define a new risk score. We show that such approach combining adult BMI and PRS achieves considerably better prediction than each of the measures on unrelated Caucasians in the UK Biobank (UKB, n = 290,584). Likewise, integrating PRS with self-reports on birth weight ( n = 172,239) and comparative body size at age ten ( n = 287,203) also substantially enhance prediction as compared to each of its components. While the integration of PRS with BMI achieved better results as compared to the other measurements, the latter are early-life measurements that can be integrated already at childhood, to allow preemptive intervention for those at high risk to develop T2D. Our integrated approach can be easily generalized to other diseases, with the relevant early-life measurements.

The routine detection, surveillance, and reporting of novel SARS-CoV-2 variants is crucial, as these threaten to hinder global vaccination efforts. Herein we report a novel local variant with a non-synonymous mutation in the spike (S) protein P681H. This local Israeli variant was not associated with a higher infection rate or higher prevalence. Furthermore, the local variant was successfully neutralized by sera from fully vaccinated individuals at a comparable level to the B.1.1.7 variant and an Israel wild-type strain. While it is not a variant of concern, routine monitoring by sequencing is still required.

Bacillus Calmette-Guerin (BCG) is a live attenuated form of Mycobacterium bovis that was developed 100 years ago as a vaccine against tuberculosis (TB) and has been used ever since to vaccinate children globally. It has also been used as the first-line treatment in patients with nonmuscle invasive bladder cancer (NMIBC), through repeated intravesical applications. Numerous studies have shown that BCG induces off-target immune effects in various pathologies. Accumulating data argue for the critical role of the immune system in the course of neurodegenerative diseases such as Alzheimer's disease (AD) and Parkinson's disease (PD). In this study, we tested whether repeated exposure to BCG during the treatment of NMIBC is associated with the risk of developing AD and PD. We presented a multi-center retrospective cohort study with patient data collected between 2000 and 2019 that included 12,185 bladder cancer (BC) patients, of which 2301 BCG-treated patients met all inclusion criteria, with a follow-up of 3.5 to 7 years. We considered the diagnosis date of AD and nonvascular dementia cases for BC patients. The BC patients were partitioned into those who underwent a transurethral resection of the bladder tumor followed by BCG therapy, and a disjoint group that had not received such treatment. By applying Cox proportional hazards (PH) regression and competing for risk analyses, we found that BCG treatment was associated with a significantly reduced risk of developing AD, especially in the population aged 75 years or older. The older population (≥75 years, 1578 BCG treated, and 5147 controls) showed a hazard ratio (HR) of 0.726 (95% CI: 0.529-0.996; p -value = 0.0473). While in a hospital-based cohort, BCG treatment resulted in an HR of 0.416 (95% CI: 0.203-0.853; p -value = 0.017), indicating a 58% lower risk of developing AD. The risk of developing PD showed the same trend with a 28% reduction in BCG-treated patients, while no BCG beneficial effect was observed for other age-related events such as Type 2 diabetes (T2D) and stroke. We attributed BCG's beneficial effect on neurodegenerative diseases to a possible activation of long-term nonspecific immune effects. We proposed a prospective study in elderly people for testing intradermic BCG inoculation as a potential protective agent against AD and PD.

Conducting numerous, rapid, and reliable PCR tests for SARS-CoV-2 is essential for our ability to monitor and control the current COVID-19 pandemic. Here, we tested the sensitivity and efficiency of SARS-CoV-2 detection in clinical samples collected directly into a mix of lysis buffer and RNA preservative, thus inactivating the virus immediately after sampling. We tested 79 COVID-19 patients and 20 healthy controls. We collected two samples (nasopharyngeal swabs) from each participant: one swab was inserted into a test tube with Viral Transport Medium (VTM), following the standard guideline used as the recommended method for sample collection; the other swab was inserted into a lysis buffer supplemented with nucleic acid stabilization mix (coined NSLB). We found that RT-qPCR tests of patients were significantly more sensitive with NSLB sampling, reaching detection threshold 2.1±0.6 (Mean±SE) PCR cycles earlier then VTM samples from the same patient. We show that this improvement is most likely since NSLB samples are not diluted in lysis buffer before RNA extraction. Re-extracting RNA from NSLB samples after 72 hours at room temperature did not affect the sensitivity of detection, demonstrating that NSLB allows for long periods of sample preservation without special cooling equipment. We also show that swirling the swab in NSLB and discarding it did not reduce sensitivity compared to retaining the swab in the tube, thus allowing improved automation of COVID-19 tests. Overall, we show that using NSLB instead of VTM can improve the sensitivity, safety, and rapidity of COVID-19 tests at a time most needed., Competing Interests: The authors have declared that no competing interests exist.

Natural language processing (NLP) is a field of computer science concerned with automated text and language analysis. In recent years, following a series of breakthroughs in deep and machine learning, NLP methods have shown overwhelming progress. Here, we review the success, promise and pitfalls of applying NLP algorithms to the study of proteins. Proteins, which can be represented as strings of amino-acid letters, are a natural fit to many NLP methods. We explore the conceptual similarities and differences between proteins and language, and review a range of protein-related tasks amenable to machine learning. We present methods for encoding the information of proteins as text and analyzing it with NLP methods, reviewing classic concepts such as bag-of-words, k-mers/n-grams and text search, as well as modern techniques such as word embedding, contextualized embedding, deep learning and neural language models. In particular, we focus on recent innovations such as masked language modeling, self-supervised learning and attention-based models. Finally, we discuss trends and challenges in the intersection of NLP and protein research., Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (© 2021 The Author(s).)

Contemporary catalogues of cancer driver genes rely primarily on high mutation rates as evidence for gene selection in tumors. Here, we present The Functional Alteration Bias Recovery In Coding-regions Cancer Portal, a comprehensive catalogue of gene selection in cancer based purely on the biochemical functional effects of mutations at the protein level. Gene selection in the portal is quantified by combining genomics data with rich proteomic annotations. Genes are ranked according to the strength of evidence for selection in tumor, based on rigorous and robust statistics. The portal covers the entire human coding genome (∼18,000 protein-coding genes) across 33 cancer types and pan-cancer. It includes a selected set of cross-references to the most relevant resources providing genomics, proteomics, and cancer-related information. We showcase the portal with known and overlooked cancer genes, demonstrating the utility of the portal via its simple visual interface, which allows users to pivot between gene-centric and cancer type views. The portal is available at fabric-cancer.huji.ac.il. SIGNIFICANCE: A new cancer portal quantifies and presents gene selection in tumor over the entire human coding genome across 33 cancer types and pan-cancer., (©2020 American Association for Cancer Research.)

MicroRNAs (miRNAs) act as negative regulators of gene expression in the cytoplasm. Previous studies have identified the presence of miRNAs in the nucleus. Here we study human breast cancer-derived cell-lines (MCF-7 and MDA-MB-231) and a non-tumorigenic cell-line (MCF-10A) and compare their miRNA sequences at the spliceosome fraction (SF). We report that the levels of miRNAs found in the spliceosome, their identity, and pre-miRNA segmental composition are cell-line specific. One such miRNA is miR-7704 whose genomic position overlaps HAGLR, a cancer-related lncRNA. We detected an inverse expression of miR-7704 and HAGLR in the tested cell lines. Specifically, inhibition of miR-7704 caused an increase in HAGLR expression. Furthermore, elevated levels of miR-7704 slightly altered the cell-cycle in MDA-MB-231. Altogether, we show that SF-miR-7704 acts as a tumor-suppressor gene with HAGLR being its nuclear target. The relative levels of miRNAs found in the spliceosome fractions (e.g., miR-100, miR-30a, and let-7 family) in non-tumorigenic relative to cancer-derived cell-lines was monitored. We found that the expression trend of the abundant miRNAs in SF was different from that reported in the literature and from the observation of large cohorts of breast cancer patients, suggesting that many SF-miRNAs act on targets that are different from the cytoplasmic ones. Altogether, we report on the potential of SF-miRNAs as an unexplored route for cancerous cell state.

Episodic and spatial memory decline in aging and are controlled by the hippocampus, perirhinal, frontal and parietal cortices and the connections between them. Ladostigil, a drug with antioxidant and anti-inflammatory activity, was shown to prevent the loss of episodic and spatial memory in aging rats. To better understand the molecular effects of aging and ladostigil on these brain regions we characterized the changes in gene expression using RNA-sequencing technology in rats aged 6 and 22 months. We found that the changes induced by aging and chronic ladostigil treatment were brain region specific. In the hippocampus, frontal and perirhinal cortex, ladostigil decreased the overexpression of genes regulating calcium homeostasis, ion channels and those adversely affecting synaptic function. In the parietal cortex, ladostigil increased the expression of several genes that provide neurotrophic support, while reducing that of pro-apoptotic genes and those encoding pro-inflammatory cytokines and their receptors. Ladostigil also decreased the expression of axonal growth inhibitors and those impairing mitochondrial function. Together, these actions could explain the protection by ladostigil against age-related memory decline., (Copyright © 2020 Elsevier Ltd. All rights reserved.)