Neuropilin-1 is a novel host factor modulating the entry of hepatitis B virus.

BACKGROUND & AIMS: Sodium taurocholate cotransporting polypeptide (NTCP) has been identified as the cellular receptor for hepatitis B virus (HBV). However, hepatocytes expressing NTCP exhibit varying susceptibilities to HBV infection. This study aimed to investigate whether other host factors modulate the process of HBV infection. METHODS: Liver biopsy samples obtained from children with hepatitis B were used for single-cell sequencing and susceptibility analysis. Primary human hepatocytes, HepG2-NTCP cells, and human liver chimeric mice were used to analyze the effect of candidate host factors on HBV infection. RESULTS: Single-cell sequencing and susceptibility analysis revealed a positive correlation between neuropilin-1 (NRP1) expression and HBV infection. In the HBV-infected cell model, NRP1 overexpression before HBV inoculation significantly enhanced viral attachment and internalization, and promoted viral infection in the presence of NTCP. Mechanistic studies indicated that NRP1 formed a complex with LHBs and NTCP. The NRP1 b domain mediated its interaction with conserved arginine residues at positions 88 and 92 in the preS1 domain of the HBV envelope protein LHBs. This NRP1-preS1 interaction subsequently promoted the binding of preS1 to NTCP, facilitating viral infection. Moreover, disruption of the NRP1-preS1 interaction by the NRP1 antagonist EG00229 significantly attenuated the binding affinity between NTCP and preS1, thereby inhibiting HBV infection both in vitro and in vivo. CONCLUSIONS: Our findings indicate that NRP1 is a novel host factor for HBV infection, which interacts with preS1 and NTCP to modulate HBV entry into hepatocytes. IMPACT AND IMPLICATIONS: HBV infection is a global public health problem, but the understanding of the early infection process of HBV remains limited. Through single-cell sequencing, we identified a novel host factor, NRP1, which modulates HBV entry by interacting with HBV preS1 and NTCP. Moreover, antagonists targeting NRP1 can inhibit HBV infection both in vitro and in vivo. This study could further advance our comprehension of the early infection process of HBV.

First copy number variant in trans with single nucleotide variant in CCN6 causing progressive pseudorheumatoid dysplasia revealed by genome sequencing and deep phenotyping in monozygotic twins.

Biallelic pathogenic variants in CCN6 cause progressive pseudorheumatoid dysplasia (PPD), a rare skeletal dysplasia. The predominant features include noninflammatory progressive joint stiffness and enlargement, which are not unique to this condition. Nearly 100% of the reported variants are single nucleotide variants or small indels, and missing of a second variant has been reported. Genome sequencing (GS) covers various types of variants and deep phenotyping (DP) provides detailed and precise information facilitating genetic data interpretation. The combination of GS and DP improves diagnostic yield, especially in rare and undiagnosed diseases. We identified a novel compound heterozygote involving a disease-causing copy number variant (g.112057664_112064205del) in trans with a single nucleotide variant (c.624dup(p.Cys209MetfsTer21)) in CCN6 in a pair of monozygotic twins, through the methods of GS and DP. The twins had received three nondiagnostic results before. The g.112057664_112064205del variant was missed by all the tests, and the recorded phenotypes were inaccurate or even misleading. The twins were diagnosed with PPD, ending a 13-year diagnostic odyssey. There may be other patients with PPD experiencing underdiagnosis and misdiagnosis due to inadequate genetic testing or phenotyping methods. This case highlights the critical role of GS and DP in facilitating an accurate and timely diagnosis.

De novo biosynthesis of ß-arbutin in Corynebacterium glutamicum via pathway engineering and process optimization.

BACKGROUND: ß-Arbutin, a hydroquinone glucoside found in pears, bearberry leaves, and various plants, exhibits antioxidant, anti-inflammatory, antimicrobial, and anticancer effects. ß-Arbutin has wide applications in the pharmaceutical and cosmetic industries. However, the limited availability of high-performance strains limits the biobased production of ß-arbutin. RESULTS: This study established the ß-arbutin biosynthetic pathway in C. glutamicum ATCC13032 by introducing codon-optimized ubiC, MNX1, and AS. Additionally, the production titer of ß-arbutin was increased by further inactivation of csm and trpE to impede the competitive metabolic pathway. Further modification of the upstream metabolic pathway and supplementation of UDP-glucose resulted in the final engineered strain, C. glutamicum AR11, which achieved a ß-arbutin production titer of 7.94 g/L in the optimized fermentation medium. CONCLUSIONS: This study represents the first successful instance of de novo ß-arbutin production in C. glutamicum, offering a chassis cell for ß-arbutin biosynthesis.

Promotion of apoptosis in melanoma cells by taxifolin through the PI3K/AKT signaling pathway: Screening of natural products using WGCNA and CMAP platforms.

Melanoma is a skin cancer originating from melanocytes. The global incidence rate of melanoma is rapidly increasing, posing significant public health challenges. Identifying effective therapeutic agents is crucial in addressing this growing problem. Natural products have demonstrated promising anti-tumor activity. In this study, a plant flavonoid, taxifolin, was screened using Weighted Correlation Network Analysis (WGCNA) in combination with the Connectivity Map (CMAP) platform. Taxifolin was confirmed to inhibit the proliferation, migration, and invasion ability of melanoma A375 and MV-3 cells by promoting apoptosis. Additionally, it suppressed the Epithelial-Mesenchymal Transition (EMT) process of melanoma cells. Cyber pharmacological analysis revealed that taxifolin exerts its inhibitory effect on melanoma through the PI3K/AKT signaling pathway, specifically by downregulating the protein expression of p-PI3K and p-AKT. Notably, the addition of SC-79, an activator of the PI3K/AKT signaling pathway, reversed the effects of taxifolin on cell migration and apoptosis. Furthermore, in vivo experiments demonstrated that taxifolin treatment slowed tumor growth in mice without significant toxic effects. Based on these findings, taxifolin holds promise as a potential drug for melanoma treatment.

A phosphorylated guanidine chitosan and UiO-66-NH2 modified magnetic nanoparticle platform for enrichment and detection of multiple bacteria.

Wastewater-based epidemiology (WBE) is a powerful tool for early warning of infectious disease outbreaks. Hence, a rapid and portable pathogen monitoring system is urgent needed for on-site detection. In this work, we first reported synthesis of an artificial modulated wide-spectrum bacteria capture nanoparticle (Arg-CSP@UiO@Fe3O4). Arginine-modified phosphorylated chitosan (Arg-CSP) coating could provide strongly positive charged guanidinium group for pathogen interaction by electrostatic attraction, and UiO-66-NH2 layer could help Arg-CSP graft onto Fe3O4 magnetic particles. The capture efficiency of Arg-CSP@UiO@Fe3O4 reached 92.2 % and 97.3 % for Escherichia coli (E.coli) and Staphylococcus epidermidis (S.epidermidis)within 40 min, in 10 mL sample. To prevent pathogen degradation in sewage, a portable nucleic acid extraction-free method was also developed. UiO-66-NH2 could disintegrate in buffer with high concentration of PO43- for bacterium desorption, and then nucleic acid of the bacteria was released by heating. The DNA template concentration obtained by this method was 779.28 times higher than that of the direct thermal lysis product and 8.63 times higher than that of the commercial kit. Afterwards, multiple detection of bacteria was realized by loop-mediated isothermal amplification (LAMP). Artificial regulated pathogen desorption could prevent non-specific adsorption of nucleic acid by nanoparticles. The detection limit of Arg-CSP@UiO@Fe3O4-LAMP method was 80 cfu/mL for E.coli and 300 cfu/mL for S.epidermidis. The accuracy and reliability of the method was validated by spiked sewage samples. In conclusion, this bio-monitoring system was able to detect multiple bacteria in environment conveniently and have good potential to become an alternative solution for rapid on-site pathogen detection.

Myostatin regulates energy homeostasis through autocrine- and paracrine-mediated microenvironment communications.

Myostatin (MSTN) has long been recognized as a critical regulator of muscle mass. Recently, there has been an increasing interest in its role in metabolism. In our study, we specifically knocked out MSTN in brown adipose tissue (BAT) from mice (MSTNΔUCP1) and found that the mice gained more weight than controls when fed a high-fat diet, with progressive hepatosteatosis and impaired skeletal muscle activity. RNA-seq analysis indicated signatures of mitochondrial dysfunction and inflammation in the MSTN-ablation BAT. Further studies demonstrated that the the Kruppel-like factor 4 (KLF4) was responsible for the metabolic phenotypes observed, while FGF21 contributed to the microenvironment communication between adipocytes and macrophages induced by the loss of MSTN. Moreover, the MSTN-SMAD2/3-p38 signaling pathway mediated the expression of KLF4 and FGF21 in adipocytes. In summary, our findings suggest that brown adipocytes-derived MSTN regulates BAT thermogenesis via autocrine and paracrine effects on adipocytes or macrophages, ultimately regulating systemic energy homeostasis.

Construction of logic gate computation for the assay of the nerve agent sarin based on an AChE-based dual-channel sensing system.

Nerve agents have posed a huge threat to national and human security, and their sensitive detection is crucial. Herein, based on the oxidation of Ce4+ and the aggregation-induced emission (AIE) of glutathione-protected gold nanoclusters (GSH-Au NCs), a cascade reaction was designed to prepare oxidized 3,3',5,5'-tetramethylbenzidine (oxTMB) and GSH-Au NCs crosslinked by Ce3+ (Ce3+-GSH-Au NCs). oxTMB had a broad UV-visible absorption range (500-700 nm) and was capable of quenching the fluorescence of Ce3+-GSH-Au NCs at 590 nm through the internal filtration effect (IFE). Thiocholine (TCh), the hydrolysis product of acetylthiocholine chloride (ATCl) catalyzed by acetylcholinesterase (AChE), reduced oxTMB completely, resulting in a decrease in the absorption of oxTMB and the recovery of IFE-quenched fluorescence of Ce3+-GSH-Au NCs. Nerve agent sarin (GB) hindered the production of TCh and the reduction of oxTMB by inhibiting the AChE activity, leading to the fluorescence of Ce3+-GSH-Au NCs being quenched again. The dual-output sensing system (AChE + ATCl + oxTMB + Ce3+-GSH-Au NCs) exhibited a low limit of detection to GB (2.46 nM for colorimetry and 1.18 nM for fluorimetry) and excellent selectivity toward common interferences being unable to inhibit AChE. Moreover, the intelligent logic gate constructed based on the sensing system showed promising applications in the field of smart sensing of nerve agents.

Orientin alleviates the inflammatory response in psoriasis like dermatitis in BALB/c mice by inhibiting the MAPK signaling pathway.

BACKGROUND: Psoriasis, a chronic inflammatory condition of the skin, is characterized by an atypical proliferation of epidermal keratinocytes and immune cell infiltration. Orientin is a flavonoid monomer with potent anti-inflammatory activities. However, the therapeutic effects of orientin on psoriasis and the underlying mechanisms have not been elucidated. OBJECTIVE: To investigate the therapeutic effect of orientin on psoriasis and the underlying mechanisms using network pharmacology and experimental studies. METHODS: A psoriasis-like mouse model was established using imiquimod (IMQ). Lipopolysaccharide (LPS) was used to stimulate the RAW264.7 and HaCaT cells in vitro. The therapeutic effects of orientin and the underlying mechanism were analyzed using histopathological, immunohistochemical, quantitative real-time polymerase chain reaction, enzyme-linked immunosorbent assay, flow cytometry, and western blotting analyses. RESULTS: Orientin ameliorated skin lesions and suppressed keratinocyte proliferation and immune cell infiltration in the IMQ-induced psoriasis-like mouse model. Additionally, orientin inhibited the secretion of the pro-inflammatory factors interleukin (IL)-1ß, tumor necrosis factor (TNF)-α, IL-6, IL-8, IL-17, and IL-23 in the psoriasis-like mouse model and LPS-induced RAW264.7 and HaCaT cells. Furthermore, orientin mitigated the LPS-induced upregulation of reactive oxygen species and downregulation of IL-10 and glutathione levels. Orientin alleviated inflammation by downregulating the MAPK signaling pathway. CONCLUSION: Orientin alleviated psoriasis-like dermatitis by suppressing the MAPK signaling pathway, suggesting that orientin is a potential therapeutic for psoriasis.

Diagnosis and treatment of the Ehlers-Danlos syndromes in China: synopsis of the first guidelines.

BACKGROUND: The Ehlers-Danlos syndromes (EDS) are a group of rare hereditary connective tissue disorders. EDS is clinically and genetically heterogeneous and usually involves multiple systems. There are 14 subtypes of EDS with hallmark features including joint hypermobility, skin hyperextensibility, and tissue fragility. The clinical manifestations and their severity differ among the subtypes, encompassing recurrent joint dislocations, scoliosis, arterial aneurysm and dissection, and organ rupture. Challenges in diagnosis and management arise from the complexity of the disease, which is further complicated by its rarity. The development of clinical guidelines and implementation of coordinated multi-disciplinary team (MDT) approaches have emerged as global priorities. MAIN BODY: Chinese Multi-Disciplinary Working Group on the Ehlers-Danlos Syndromes was therefore established. Healthcare professionals were recruited from 25 top hospitals across China. The experts are specialized in 24 fields, including genetics, vascular surgery, dermatology, and orthopedics, as well as nursing care, rehabilitation, psychology, and nutrition. Based on GRADE methodology, the Guidelines were written by the Group supervised by methodologists, following a systemic review of all 4453 articles in PubMed published before August 9, 2023, using the search term "Ehlers Danlos". A coordinated MDT approach for the diagnosis and management of EDS is highly recommended by the Group, along with 29 specific recommendations addressing key clinical questions. In addition to the treatment plan, the Guidelines also emphasize integrating support from nursing care, rehabilitation, psychology, and nutrition. This integration not only facilitates recovery in hospital settings, but most importantly, the transition from an illness-defined life to a more "normalized" life. CONCLUSION: The first guidelines on EDS will shorten the diagnostic odyssey and solve the unmet medical needs of the patients. This article is a synopsis of the full guidelines.

Von Hippel Lindau tumor suppressor controls m6A-dependent gene expression in renal tumorigenesis.

N6-Methyladenosine (m6A) is the most abundant posttranscriptional modification, and its contribution to cancer evolution has recently been appreciated. Renal cancer is the most common adult genitourinary cancer, approximately 85% of which is accounted for by the clear cell renal cell carcinoma (ccRCC) subtype characterized by VHL loss. However, it is unclear whether VHL loss in ccRCC affects m6A patterns. In this study, we demonstrate that VHL binds and promotes METTL3/METTL14 complex formation while VHL depletion suppresses m6A modification, which is distinctive from its canonical E3 ligase role. m6A RNA immunoprecipitation sequencing (RIP-Seq) coupled with RNA-Seq allows us to identify a selection of genes whose expression may be regulated by VHL-m6A signaling. Specifically, PIK3R3 is identified to be a critical gene whose mRNA stability is regulated by VHL in a m6A-dependent but HIF-independent manner. Functionally, PIK3R3 depletion promotes renal cancer cell growth and orthotopic tumor growth while its overexpression leads to decreased tumorigenesis. Mechanistically, the VHL-m6A-regulated PIK3R3 suppresses tumor growth by restraining PI3K/AKT activity. Taken together, we propose a mechanism by which VHL regulates m6A through modulation of METTL3/METTL14 complex formation, thereby promoting PIK3R3 mRNA stability and protein levels that are critical for regulating ccRCC tumorigenesis.

Core planar cell polarity genes VANGL1 and VANGL2 in predisposition to congenital vertebral malformations.

Congenital scoliosis (CS), affecting approximately 0.5 to 1 in 1,000 live births, is commonly caused by congenital vertebral malformations (CVMs) arising from aberrant somitogenesis or somite differentiation. While Wnt/ß-catenin signaling has been implicated in somite development, the function of Wnt/planar cell polarity (Wnt/PCP) signaling in this process remains unclear. Here, we investigated the role of Vangl1 and Vangl2 in vertebral development and found that their deletion causes vertebral anomalies resembling human CVMs. Analysis of exome sequencing data from multiethnic CS patients revealed a number of rare and deleterious variants in VANGL1 and VANGL2, many of which exhibited loss-of-function and dominant-negative effects. Zebrafish models confirmed the pathogenicity of these variants. Furthermore, we found that Vangl1 knock-in (p.R258H) mice exhibited vertebral malformations in a Vangl gene dose- and environment-dependent manner. Our findings highlight critical roles for PCP signaling in vertebral development and predisposition to CVMs in CS patients, providing insights into the molecular mechanisms underlying this disorder.

[Research progress on emotion recognition by combining virtual reality environment and electroencephalogram signals].

Emotion recognition refers to the process of determining and identifying an individual's current emotional state by analyzing various signals such as voice, facial expressions, and physiological indicators etc. Using electroencephalogram (EEG) signals and virtual reality (VR) technology for emotion recognition research helps to better understand human emotional changes, enabling applications in areas such as psychological therapy, education, and training to enhance people's quality of life. However, there is a lack of comprehensive review literature summarizing the combined researches of EEG signals and VR environments for emotion recognition. Therefore, this paper summarizes and synthesizes relevant research from the past five years. Firstly, it introduces the relevant theories of VR and EEG signal emotion recognition. Secondly, it focuses on the analysis of emotion induction, feature extraction, and classification methods in emotion recognition using EEG signals within VR environments. The article concludes by summarizing the research's application directions and providing an outlook on future development trends, aiming to serve as a reference for researchers in related fields.

Metal-organic polymer enables efficient organic photoelectrochemical transistor biosensing.

The field of organic photoelectrochemical transistor (OPECT) is newly emerged, with increasing efforts attempting to utilize its properties in biological sensing. Advanced materials with new physicochemical properties have proven important to this end. Herein, we report a metal-organic polymers-gated OPECT biosensing exemplified by CuⅠ-arylacetylide polymers (CuAs)-modulated poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) channel. Both the photoelectrochemical properties and gating capability of CuAs are explored and optimized for high-efficacy photogating. Morever, based on its inherent structure, the specific reaction between CuAs and sulfur ions (S2-) is revealed and S2--mediated microRNA-21 detection is realized by linking with nucleic acid amplification and alkaline phosphatase catalytic chemistry. This work introduces metal-organic polymers as gating materials for OPECT biosensing.

Kinesin Family Member-18A (KIF18A) Promotes Cell Proliferation and Metastasis in Hepatocellular Carcinoma.

BACKGROUND & AIMS: Kinesin family member 18A (KIF18A) is notable for its aberrant expression across various cancer types and its pivotal role is driving cancer progression. In this study, we aim to investigate the intricate molecular mechanisms underlying the impact of KIF18A on the progression of HCC. METHODS: Western blotting assays, a quantitative real-time PCR and immunohistochemical analyses were performed to quantitatively assess KIF18A expression in HCC tissues. We then performed genetic manipulations within HCC cells by silencing endogenous KIF18A using short hairpin RNA (shRNA) and introducing exogenous plasmids to overexpress KIF18A. We monitored cell progression, analyzed cell cycle and cell apoptosis and assessed cell migration and invasion both in vitro and in vivo. Moreover, we conducted RNA-sequencing to explore KIF18A-related signaling pathways utilizing Reactome and KEGG enrichment methods and validated these critical mediators in these pathways. RESULTS: Analysis of the TCGA-LIHC database revealed pronounced overexpression of KIF18A in HCC tissues, the finding was subsequently confirmed through the analysis of clinical samples obtained from HCC patients. Notably, silencing KIF18A in cells led to an obvious inhibition of cell proliferation, migration and invasion in vitro. Furthermore, in subcutaneous and orthotopic xenograft models, suppression of KIF18A sgnificantly redudce tumor weight and the number of lung metastatic nodules. Mechanistically, KIF18A appears to facilitate cell proliferation by upregulating MAD2 and CDK1/CyclinB1 expression levels, with the activation of SMAD2/3 signaling contributing to KIF18A-driven metastasis. CONCLUSION: Our study elucidates the molecular mechanism by which KIF18A mediates proliferation and metastasis in HCC cells, offering new insights into potential therapeutic targets.

Nocardia rubra cell-wall skeleton mitigates whole abdominal irradiation-induced intestinal injury via regulating macrophage function.

Background: Ionizing radiation (IR)-induced intestinal injury is a major side effect and dose-limiting toxicity in patients receiving radiotherapy. There is an urgent need to identify an effective and safe radioprotectant to reduce radiation-induced intestinal injury. Immunoregulation is considered an effective strategy against IR-induced injury. The purpose of this article was to investigate the protective effect of Nocardia rubra cell wall skeleton (Nr-CWS), an immunomodulator, on radiation-induced intestinal damage and to explore its potential mechanism. Methods: C57BL/6 J male mice exposed to 12 Gy whole abdominal irradiation (WAI) were examined for survival rate, morphology and function of the intestine and spleen, as well as the gut microbiota, to comprehensively evaluate the therapeutic effects of Nr-CWS on radiation-induced intestinal and splenetic injury. To further elucidate the underlying mechanisms of Nr-CWS-mediated intestinal protection, macrophages were depleted by clodronate liposomes to determine whether Nr-CWS-induced radioprotection is macrophage dependent, and the function of peritoneal macrophages stimulated by Nr-CWS was detected in vitro. Results: Our data showed that Nr-CWS promoted the recovery of intestinal barrier function, enhanced leucine-rich repeat-containing G protein-coupled receptor 5+ intestinal stem cell survival and the regeneration of intestinal epithelial cells, maintained intestinal flora homeostasis, protected spleen morphology and function, and improved the outcome of mice exposed to 12 Gy WAI. Mechanistic studies indicated that Nr-CWS recruited macrophages to reduce WAI-induced intestinal damage. Moreover, macrophage depletion by clodronate liposomes blocked Nr-CWS-induced radioprotection. In vitro, we found that Nr-CWS activated the nuclear factor kappa-B signaling pathway and promoted the phagocytosis and migration ability of peritoneal macrophages. Conclusions: Our study suggests the therapeutic effect of Nr-CWS on radiation-induced intestinal injury, and provides possible therapeutic strategy and potential preventive and therapeutic drugs to alleviate it.

Amplification-Free Analysis of Bladder Cancer MicroRNAs on Wrinkled Silica Nanoparticles with DNA-Functionalized Quantum Dots.

Bladder cancer (BC) occurrence and progression are accompanied by alterations in microRNAs (miRNAs) expression levels. Simultaneous detection of multiple miRNAs contributes to the accuracy and reliability of the BC diagnosis. In this work, wrinkled silica nanoparticles (WSNs) were applied as the microreactor for multiplex miRNAs analysis without enzymes or nucleic acid amplification. Conjugated on the surface of WSNs, the S9.6 antibody was adopted as the universal module for binding DNA/miRNA duplexes, regardless of their sequence. Furthermore, single-stranded DNA (ssDNA) was labeled with quantum dots (QDs) for identifying a given miRNA to form QDs-ssDNA/miRNA, which enabled the specific capture of the corresponding QDs on the wrinkled surface of WSNs. Based on the detection of fluorescence signals that were ultimately focused on WSNs, target miRNAs could be sensitively identified to a femtomolar level (5 fM) with a wide dynamic range of up to 6 orders of magnitude. The proposed strategy achieved high specificity to obviously distinguish single-base mutation sequences and possessed multiplex assay capability. Moreover, the assay exhibited excellent practicability in the multiplex detection of miRNAs in clinical serum specimens.

Novel methods for the rapid and sensitive detection of Nipah virus based on a CRISPR/Cas12a system.

Nipah virus (NiV), a bat-borne zoonotic viral pathogen with high infectivity and lethality to humans, has caused severe outbreaks in several countries of Asia during the past two decades. Because of the worldwide distribution of the NiV natural reservoir, fruit bats, and lack of effective treatments or vaccines for NiV, routine surveillance and early detection are the key measures for containing NiV outbreaks and reducing its influence. In this study, we developed two rapid, sensitive and easy-to-conduct methods, RAA-CRISPR/Cas12a-FQ and RAA-CRISPR/Cas12a-FB, for NiV detection based on a recombinase-aided amplification (RAA) assay and a CRISPR/Cas12a system by utilizing dual-labeled fluorophore-quencher or fluorophore-biotin ssDNA probes. These two methods can be completed in 45 min and 55 min and achieve a limit of detection of 10 copies per µL and 100 copies per µL of NiV N DNA, respectively. In addition, they do not cross-react with nontarget nucleic acids extracted from the pathogens causing similar symptoms to NiV, showing high specificity for NiV N DNA detection. Meanwhile, they show satisfactory performance in the detection of spiked samples from pigs and humans. Collectively, the RAA-CRISPR/Cas12a-FQ and RAA-CRISPR/Cas12a-FB methods developed by us would be promising candidates for the early detection and routine surveillance of NiV in resource-poor areas and outdoors.

LAVASET: Latent Variable Stochastic Ensemble of Trees. An ensemble method for correlated datasets with spatial, spectral, and temporal dependencies.

MOTIVATION: Random forests (RFs) can deal with a large number of variables, achieve reasonable prediction scores, and yield highly interpretable feature importance values. As such, RFs are appropriate models for feature selection and further dimension reduction. However, RFs are often not appropriate for correlated datasets due to their mode of selecting individual features for splitting. Addressing correlation relationships in high-dimensional datasets is imperative for reducing the number of variables that are assigned high importance, hence making the dimension reduction most efficient. Here, we propose the LAtent VAriable Stochastic Ensemble of Trees (LAVASET) method that derives latent variables based on the distance characteristics of each feature and aims to incorporate the correlation factor in the splitting step. RESULTS: Without compromising on performance in the majority of examples, LAVASET outperforms RF by accurately determining feature importance across all correlated variables and ensuring proper distribution of importance values. LAVASET yields mostly non-inferior prediction accuracies to traditional RFs when tested in simulated and real 1D datasets, as well as more complex and high-dimensional 3D datatypes. Unlike traditional RFs, LAVASET is unaffected by single 'important' noisy features (false positives), as it considers the local neighbourhood. LAVASET, therefore, highlights neighbourhoods of features, reflecting real signals that collectively impact the model's predictive ability. AVAILABILITY AND IMPLEMENTATION: LAVASET is freely available as a standalone package from https://github.com/melkasapi/LAVASET.