Multi-organ proteomic landscape of COVID-19 autopsies.

The molecular pathology of multi-organ injuries in COVID-19 patients remains unclear, preventing effective therapeutics development. Here, we report a proteomic analysis of 144 autopsy samples from seven organs in 19 COVID-19 patients. We quantified 11,394 proteins in these samples, in which 5,336 were perturbed in the COVID-19 patients compared to controls. Our data showed that cathepsin L1, rather than ACE2, was significantly upregulated in the lung from the COVID-19 patients. Systemic hyperinflammation and dysregulation of glucose and fatty acid metabolism were detected in multiple organs. We also observed dysregulation of key factors involved in hypoxia, angiogenesis, blood coagulation, and fibrosis in multiple organs from the COVID-19 patients. Evidence for testicular injuries includes reduced Leydig cells, suppressed cholesterol biosynthesis, and sperm mobility. In summary, this study depicts a multi-organ proteomic landscape of COVID-19 autopsies that furthers our understanding of the biological basis of COVID-19 pathology.

Neuroimmune cardiovascular interfaces control atherosclerosis.

Atherosclerotic plaques develop in the inner intimal layer of arteries and can cause heart attacks and strokes1. As plaques lack innervation, the effects of neuronal control on atherosclerosis remain unclear. However, the immune system responds to plaques by forming leukocyte infiltrates in the outer connective tissue coat of arteries (the adventitia)2-6. Here, because the peripheral nervous system uses the adventitia as its principal conduit to reach distant targets7-9, we postulated that the peripheral nervous system may directly interact with diseased arteries. Unexpectedly, widespread neuroimmune cardiovascular interfaces (NICIs) arose in mouse and human atherosclerosis-diseased adventitia segments showed expanded axon networks, including growth cones at axon endings near immune cells and media smooth muscle cells. Mouse NICIs established a structural artery-brain circuit (ABC): abdominal adventitia nociceptive afferents10-14 entered the central nervous system through spinal cord T6-T13 dorsal root ganglia and were traced to higher brain regions, including the parabrachial and central amygdala neurons; and sympathetic efferent neurons projected from medullary and hypothalamic neurons to the adventitia through spinal intermediolateral neurons and both coeliac and sympathetic chain ganglia. Moreover, ABC peripheral nervous system components were activated: splenic sympathetic and coeliac vagus nerve activities increased in parallel to disease progression, whereas coeliac ganglionectomy led to the disintegration of adventitial NICIs, reduced disease progression and enhanced plaque stability. Thus, the peripheral nervous system uses NICIs to assemble a structural ABC, and therapeutic intervention in the ABC attenuates atherosclerosis.

Molecular basis of locus-specific H3K9 methylation catalyzed by SUVH6 in plants.

Dimethylated histone H3 Lys9 (H3K9me2) is a conserved heterochromatic mark catalyzed by SUPPRESSOR OF VARIEGATION 3-9 HOMOLOG (SUVH) methyltransferases in plants. However, the mechanism underlying the locus specificity of SUVH enzymes has long been elusive. Here, we show that a conserved N-terminal motif is essential for SUVH6-mediated H3K9me2 deposition in planta. The SUVH6 N-terminal peptide can be recognized by the bromo-adjacent homology (BAH) domain of the RNA- and chromatin-binding protein ANTI-SILENCING 1 (ASI1), which has been shown to function in a complex to confer gene expression regulation. Structural data indicate that a classic aromatic cage of ASI1-BAH domain specifically recognizes an arginine residue of SUVH6 through extensive hydrogen bonding interactions. A classic aromatic cage of ASI1 specifically recognizes an arginine residue of SUVH6 through extensive cation-π interactions, playing a key role in recognition. The SUVH6-ASI1 module confers locus-specific H3K9me2 deposition at most SUVH6 target loci and gives rise to distinct regulation of gene expression depending on the target loci, either conferring transcriptional silencing or posttranscriptional processing of mRNA. More importantly, such mechanism is conserved in multiple plant species, indicating a coordinated evolutionary process between SUVH6 and ASI1. In summary, our findings uncover a conserved mechanism for the locus specificity of H3K9 methylation in planta. These findings provide mechanistic insights into the delicate regulation of H3K9 methylation homeostasis in plants.

MPCLCDA: predicting circRNA-disease associations by using automatically selected meta-path and contrastive learning.

Circular RNA (circRNA) is closely associated with human diseases. Accordingly, identifying the associations between human diseases and circRNA can help in disease prevention, diagnosis and treatment. Traditional methods are time consuming and laborious. Meanwhile, computational models can effectively predict potential circRNA-disease associations (CDAs), but are restricted by limited data, resulting in data with high dimension and imbalance. In this study, we propose a model based on automatically selected meta-path and contrastive learning, called the MPCLCDA model. First, the model constructs a new heterogeneous network based on circRNA similarity, disease similarity and known association, via automatically selected meta-path and obtains the low-dimensional fusion features of nodes via graph convolutional networks. Then, contrastive learning is used to optimize the fusion features further, and obtain the node features that make the distinction between positive and negative samples more evident. Finally, circRNA-disease scores are predicted through a multilayer perceptron. The proposed method is compared with advanced methods on four datasets. The average area under the receiver operating characteristic curve, area under the precision-recall curve and F1 score under 5-fold cross-validation reached 0.9752, 0.9831 and 0.9745, respectively. Simultaneously, case studies on human diseases further prove the predictive ability and application value of this method.

NSRGRN: a network structure refinement method for gene regulatory network inference.

The elucidation of gene regulatory networks (GRNs) is one of the central challenges of systems biology, which is crucial for understanding pathogenesis and curing diseases. Various computational methods have been developed for GRN inference, but identifying redundant regulation remains a fundamental problem. Although considering topological properties and edge importance measures simultaneously can identify and reduce redundant regulations, how to address their respective weaknesses whilst leveraging their strengths is a critical problem faced by researchers. Here, we propose a network structure refinement method for GRN (NSRGRN) that effectively combines the topological properties and edge importance measures during GRN inference. NSRGRN has two major parts. The first part constructs a preliminary ranking list of gene regulations to avoid starting the GRN inference from a directed complete graph. The second part develops a novel network structure refinement (NSR) algorithm to refine the network structure from local and global topology perspectives. Specifically, the Conditional Mutual Information with Directionality and network motifs are applied to optimise the local topology, and the lower and upper networks are used to balance the bilateral relationship between the local topology's optimisation and the global topology's maintenance. NSRGRN is compared with six state-of-the-art methods on three datasets (26 networks in total), and it shows the best all-round performance. Furthermore, when acting as a post-processing step, the NSR algorithm can improve the results of other methods in most datasets.

Deciphering a crucial dimeric interface governing Norrin dimerization and the pathogenesis of familial exudative vitreoretinopathy.

Familial exudative vitreoretinopathy (FEVR) is a hereditary eye disease that could cause blindness. It has been established that Norrin forms dimers to activate ß-catenin signaling, yet the core interface for Norrin dimerization and the precise mechanism by which Norrin dimerization contributes to the pathogenesis of FEVR remain elusive. Here, we report an NDP variant, c.265T>C (p.Phe89Leu), that interrupted ß-catenin signaling by disrupting Norrin dimerization. Structural and functional analysis revealed that the Phe-89 of one Norrin monomer interacts with Pro-98, Ser-101, Arg-121, and Ile-123 of another, forming two core symmetrical dimerization interfaces that are pivotal for the formation of a "hand-by-arm" dimer. Intriguingly, we proved that one of the two core symmetrical interfaces is sufficient for dimerization and activation of ß-catenin signaling, with a substantial contribution from the Phe-89/Pro-98 interaction. Further functional analysis revealed that the disruption of both dimeric interfaces eliminates potential binding sites for LRP5, which could be partially restored by over-expression of TSPAN12. In conclusion, our findings unveil a core dimerization interface that regulates Norrin/LRP5 interaction, highlighting the essential role of Norrin dimerization on ß-catenin signaling and providing potential therapeutic avenues for the treatment of FEVR.

The solution structure of the ADAR2 dsRBM-RNA complex reveals a sequence-specific readout of the minor groove.

Sequence-dependent recognition of dsDNA-binding proteins is well understood, yet sequence-specific recognition of dsRNA by proteins remains largely unknown, despite their importance in RNA maturation pathways. Adenosine deaminases that act on RNA (ADARs) recode genomic information by the site-selective deamination of adenosine. Here, we report the solution structure of the ADAR2 double-stranded RNA-binding motifs (dsRBMs) bound to a stem-loop pre-mRNA encoding the R/G editing site of GluR-2. The structure provides a molecular basis for how dsRBMs recognize the shape, and also more surprisingly, the sequence of the dsRNA. The unexpected direct readout of the RNA primary sequence by dsRBMs is achieved via the minor groove of the dsRNA and this recognition is critical for both editing and binding affinity at the R/G site of GluR-2. More generally, our findings suggest a solution to the sequence-specific paradox faced by many dsRBM-containing proteins that are involved in post-transcriptional regulation of gene expression.

Metabolic reprogramming directed by super-enhancers in tumors: An emerging landscape.

Metabolic reprogramming is an essential hallmark of tumors, and metabolic abnormalities are strongly associated with the malignant phenotype of tumor cells. This is closely related to transcriptional dysregulation. Super-enhancers are extremely active cis-regulatory regions in the genome, and can amalgamate a complex set of transcriptional regulatory components that are crucial for establishing tumor cell identity, promoting tumorigenesis, and enhancing aggressiveness. In addition, alterations in metabolic signaling pathways are often accompanied by changes in super-enhancers. Presently, there is a surge in interest in the potential pathogenesis of various tumors through the transcriptional regulation of super-enhancers and oncogenic mutations in super-enhancers. In this review, we summarize the functions of super-enhancers, oncogenic signaling pathways, and tumor metabolic reprogramming. In particular, we focus on the role of the super-enhancer in tumor metabolism and its impact on metabolic reprogramming. This review also discusses the prospects and directions in the field of super-enhancer and metabolic reprogramming.

Proteome Landscapes of Human Hepatocellular Carcinoma and Intrahepatic Cholangiocarcinoma.

Liver cancer is among the top leading causes of cancer mortality worldwide. Particularly, hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (CCA) have been extensively investigated from the aspect of tumor biology. However, a comprehensive and systematic understanding of the molecular characteristics of HCC and CCA remains absent. Here, we characterized the proteome landscapes of HCC and CCA using the data-independent acquisition (DIA) mass spectrometry (MS) method. By comparing the quantitative proteomes of HCC and CCA, we found several differences between the two cancer types. In particular, we found an abnormal lipid metabolism in HCC and activated extracellular matrix-related pathways in CCA. We next developed a three-protein classifier to distinguish CCA from HCC, achieving an area under the curve (AUC) of 0.92, and an accuracy of 90% in an independent validation cohort of 51 patients. The distinct molecular characteristics of HCC and CCA presented in this study provide new insights into the tumor biology of these two major important primary liver cancers. Our findings may help develop more efficient diagnostic approaches and new targeted drug treatments.

FerrDb V2: update of the manually curated database of ferroptosis regulators and ferroptosis-disease associations.

Ferroptosis is a mode of regulated cell death characterized by iron-dependent accumulation of lipid peroxidation. It is closely linked to the pathophysiological processes in many diseases. Since our publication of the first ferroptosis database in 2020 (FerrDb V1), many new findings have been published. To keep up with the rapid progress in ferroptosis research and to provide timely and high-quality data, here we present the successor, FerrDb V2. It contains 1001 ferroptosis regulators and 143 ferroptosis-disease associations manually curated from 3288 articles. Specifically, there are 621 gene regulators, of which 264 are drivers, 238 are suppressors, 9 are markers, and 110 are unclassified genes; and there are 380 substance regulators, with 201 inducers and 179 inhibitors. Compared to FerrDb V1, curated articles increase by >300%, ferroptosis regulators increase by 175%, and ferroptosis-disease associations increase by 50.5%. Circular RNA and pseudogene are novel regulators in FerrDb V2, and the percentage of non-coding RNA increases from 7.3% to 13.6%. External gene-related data were integrated, enabling thought-provoking and gene-oriented analysis in FerrDb V2. In conclusion, FerrDb V2 will help to acquire deeper insights into ferroptosis. FerrDb V2 is freely accessible at http://www.zhounan.org/ferrdb/.

Transcriptome-wide subtyping of pediatric and adult T cell acute lymphoblastic leukemia in an international study of 707 cases.

T cell acute lymphoblastic leukemia (T-ALL) is an aggressive hematological malignancy of T cell progenitors, known to be a heterogeneous disease in pediatric and adult patients. Here we attempted to better understand the disease at the molecular level based on the transcriptomic landscape of 707 T-ALL patients (510 pediatric, 190 adult patients, and 7 with unknown age; 599 from published cohorts and 108 newly investigated). Leveraging the information of gene expression enabled us to identify 10 subtypes (G1­G10), including the previously undescribed one characterized by GATA3 mutations, with GATA3R276Q capable of affecting lymphocyte development in zebrafish. Through associating with T cell differentiation stages, we found that high expression of LYL1/LMO2/SPI1/HOXA (G1­G6) might represent the early T cell progenitor, pro/precortical/cortical stage with a relatively high age of disease onset, and lymphoblasts with TLX3/TLX1 high expression (G7­G8) could be blocked at the cortical/postcortical stage, while those with high expression of NKX2-1/TAL1/LMO1 (G9­G10) might correspond to cortical/postcortical/mature stages of T cell development. Notably, adult patients harbored more cooperative mutations among epigenetic regulators, and genes involved in JAK-STAT and RAS signaling pathways, with 44% of patients aged 40 y or above in G1 bearing DNMT3A/IDH2 mutations usually seen in acute myeloid leukemia, suggesting the nature of mixed phenotype acute leukemia.

The role of sirtuins in intervertebral disc degeneration: Mechanisms and therapeutic potential.

Intervertebral disc degeneration (IDD) is one of the main causes of low back pain, which affects the patients' quality of life and health and imposes a significant socioeconomic burden. Despite great efforts made by researchers to understand the pathogenesis of IDD, effective strategies for preventing and treating this disease remain very limited. Sirtuins are a highly conserved family of (NAD+)-dependent deacetylases in mammals that are involved in a variety of metabolic processes in vivo. In recent years, sirtuins have attracted much attention owing to their regulatory roles in IDD on physiological activities such as inflammation, apoptosis, autophagy, aging, oxidative stress, and mitochondrial function. At the same time, many studies have explored the therapeutic effects of sirtuins-targeting activators or micro-RNA in IDD. This review summarizes the molecular pathways of sirtuins involved in IDD, and summarizes the therapeutic role of activators or micro-RNA targeting Sirtuins in IDD, as well as the current limitations and challenges, with a view to provide possible solutions for the treatment of IDD.

STK16 promoted colorectal cancer progress in a c-MYC signaling-dependent manner.

BACKGROUND: Colorectal cancer standed as a global health challenge, ranking third in cancer incidence and second in cancer-related deaths worldwide. A deeper understanding of the intricate mechanisms driving colorectal cancer development was pressing need. STK16 had garnered attention in recent researches, while its involvement in cancer had been minimally explored. c-MYC had emerged as a key player in cancer biology. Due to its complex structure, multifunctionality, and intricate interactions, directly inhibiting the activity of c-MYC proves to be challenging. Hence, current research was directing efforts towards modulating c-MYC expression levels. METHODS: Immunoblot, Immunohistochemistry and immunoprecipitation assays were conducted to assess the indicated protein expression levels. RT-PCR was performed to detect the corresponding mRNA expression levels. The proliferation, migration, invasion, and colony formation abilities of the specified cancer cells were investigated using CCK8 assays, Brdu assays, transwell assays, and colony formation assays, respectively. Cellular and animal experiments were performed to investigate the correlation between STK16 signaling and c-MYC signaling. RESULTS: STK16 plays a positive regulatory role in the progression of colorectal cancer. Delving into the molecular mechanisms, we unveiled that STK16 phosphorylated c-MYC at serine 452, a pivotal event hindering the ubiquitin-proteasome pathway degradation of c-MYC. Importantly, colorectal cancer proliferation mediated by STK16 was found to be dependent on the phosphorylation of c-MYC at S452. Furthermore, the researchers demonstrated that STK16 knockout or pharmacological inhibition significantly curtailed colorectal cancer proliferation and c-MYC expression in in vivo animal models. CONCLUSION: We discovered that STK16 phosphorylates c-MYC at serine 452, hindering its degradation via the ubiquitin-proteasome pathway. STK16 inhibition, either genetically or pharmacologically, effectively curtails cancer growth and c-MYC expression in vivo. These findings highlight STK16 as a potential therapeutic target for colorectal cancer.

Characterization of a novel heterozygous frameshift variant in NDP gene that causes familial exudative vitreoretinopathy in female patients.

Familial exudative vitreoretinopathy (FEVR) is a severe inherited disease characterized by defective retinal vascular development. With genetic and clinical heterogeneity, FEVR can be inherited in different patterns and characterized by phenotypes ranging from moderate visual defects to complete vision loss. This study was conducted to unravel the genetic and functional etiology of a 4-month-old female FEVR patient. Targeted gene panel and Sanger sequencing were utilized for genetic evaluation. Luciferase assays, western blot, quantitive real-time PCR, and immunocytochemistry were performed to verify the functional defects in the identified candidate variant. Here, we report a 4-month-old girl with bilateral retinal folds and peripheral avascularization, and identified a novel frameshift heterozygous variant c.37dup (p.Leu13ProfsTer13) in NDP. In vitro experiments revealed that the Leu13ProfsTer13 variant led to a prominent decrease in protein levels instead of mRNA levels, resulting in compromised Norrin/ß-catenin signaling activity. Human androgen receptor assay further revealed that a slight skewing of X chromosome inactivation could partially cause FEVR. Thus, the pathogenic mechanism by which heterozygous frameshift or nonsense variants in female carriers cause FEVR might largely result from a loss-of-function variant in one X chromosome allele and a slightly skewed X-inactivation. Further recruitment of more FEVR-affected females carrying NDP variants and genotype-phenotype correlation analysis can ultimately offer valuable information for the prognosis prediction of FEVR.

Microwave Ablation versus Surgical Resection for US-detected Multifocal T1N0M0 Papillary Thyroid Carcinoma: A 10-Center Study.

Background Microwave ablation (MWA) is currently under preliminary investigation for the treatment of multifocal papillary thyroid carcinoma (PTC) and has shown promising treatment efficacy. Compared with surgical resection (SR), MWA is minimally invasive and could preserve thyroid function. However, a comparative analysis between MWA and SR is warranted to draw definitive conclusions. Purpose To compare MWA and SR for preoperative US-detected T1N0M0 multifocal PTC in terms of overall and 1-, 3-, and 5-year progression-free survival rates and complication rates. Materials and Methods In this retrospective study, 775 patients with preoperative US-detected T1N0M0 multifocal PTC treated with MWA or SR across 10 centers between May 2015 and December 2021 were included. Propensity score matching (PSM) was performed for patients in the MWA and SR groups, followed by comparisons between the two groups. The primary outcomes were overall and 1-, 3-, and 5-year progression-free survival (PFS) rates and complication rates. Results After PSM, 229 patients (median age, 44 years [IQR 36.5-50.5 years]; 179 female) in the MWA group and 453 patients (median age, 45 years [IQR 37-53 years]; 367 female) in the SR group were observed for a median of 20 months (range, 12-74 months) and 26 months (range, 12-64 months), respectively. MWA resulted in less blood loss, shorter incision length, and shorter procedure and hospitalization durations (all P < .001). There was no evidence of differences in overall and 1-, 3-, or 5-year PFS rates (all P > .05) between MWA and SR (5-year rate, 77.2% vs 83.1%; P = .36) groups. Permanent hoarseness (2.2%, P = .05) and hypoparathyroidism (4.0%, P = .005) were encountered only in the SR group. Conclusion There was no evidence of a significant difference in PFS rates between MWA and SR for US-detected multifocal T1N0M0 PTC, and MWA resulted in fewer complications. Therefore, MWA is a feasible option for selected patients with multifocal T1N0M0 PTC. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Georgiades in this issue.

Mitochondrial stress: a key role of neuroinflammation in stroke.

Stroke is a clinical syndrome characterized by an acute, focal neurological deficit, primarily caused by the occlusion or rupture of cerebral blood vessels. In stroke, neuroinflammation emerges as a pivotal event contributing to neuronal cell death. The occurrence and progression of neuroinflammation entail intricate processes, prominently featuring mitochondrial dysfunction and adaptive responses. Mitochondria, a double membrane-bound organelle are recognized as the "energy workshop" of the body. Brain is particularly vulnerable to mitochondrial disturbances due to its high energy demands from mitochondria-related energy production. The interplay between mitochondria and neuroinflammation plays a significant role in the pathogenesis of stroke. The biological and pathological consequences resulting from mitochondrial stress have substantial implications for cerebral function. Mitochondrial stress serves as an adaptive mechanism aimed at mitigating the stress induced by the import of misfolded proteins, which occurs in response to stroke. This adaptive response involves a reduction in misfolded protein accumulation and overall protein synthesis. The influence of mitochondrial stress on the pathological state of stroke is underscored by its capacity to interact with neuroinflammation. The impact of mitochondrial stress on neuroinflammation varies according to its severity. Moderate mitochondrial stress can bolster cellular adaptive defenses, enabling cells to better withstand detrimental stressors. In contrast, sustained and excessive mitochondrial stress detrimentally affects cellular and tissue integrity. The relationship between neuroinflammation and mitochondrial stress depends on the degree of mitochondrial stress present. Understanding its role in stroke pathogenesis is instrumental in excavating the novel treatment of stroke. This review aims to provide the evaluation of the cross-talk between mitochondrial stress and neuroinflammation within the context of stroke. We aim to reveal how mitochondrial stress affects neuroinflammation environment in stroke.

Reversible Switching Between Microwave Absorption and EMI Shielding of VO2 Composite Foam.

Developing lightweight composite with reversible switching between microwave (MW) absorption and electromagnetic interference (EMI) shielding is promising yet remains highly challenging due to the completely inconsistent attenuation mechanism for electromagnetic (EM) radiation. Here, a lightweight vanadium dioxide/expanded polymer microsphere composites foam (VO2/EPM) is designed and fabricated with porous structures and 3D VO2 interconnection, which possesses reversible switching function between MW absorption and EMI shielding under thermal stimulation. The VO2/EPM exhibits MW absorption with a broad effective absorption bandwidth of 3.25 GHz at room temperature (25 °C), while provides EMI shielding of 23.1 dB at moderately high temperature (100 °C). This reversible switching performance relies on the porous structure and tunability of electrical conductivity, complex permittivity, and impedance matching, which are substantially induced by the convertible crystal structure and electronic structure of VO2. Finite element simulation is employed to qualitatively investigate the change in interaction between EM waves and VO2/EPM before and after the phase transition. Moreover, the application of VO2/EPM is demonstrated with a reversible switching function in controlling wireless transmission on/off, showcasing its excellent cycling stability. This kind of smart material with a reversible switching function shows great potential in next-generation electronic devices.

RNMFLP: Predicting circRNA-disease associations based on robust nonnegative matrix factorization and label propagation.

Circular RNAs (circRNAs) are a class of structurally stable endogenous noncoding RNA molecules. Increasing studies indicate that circRNAs play vital roles in human diseases. However, validating disease-related circRNAs in vivo is costly and time-consuming. A reliable and effective computational method to identify circRNA-disease associations deserves further studies. In this study, we propose a computational method called RNMFLP that combines robust nonnegative matrix factorization (RNMF) and label propagation algorithm (LP) to predict circRNA-disease associations. First, to reduce the impact of false negative data, the original circRNA-disease adjacency matrix is updated by matrix multiplication using the integrated circRNA similarity and the disease similarity information. Subsequently, the RNMF algorithm is used to obtain the restricted latent space to capture potential circRNA-disease pairs from the association matrix. Finally, the LP algorithm is utilized to predict more accurate circRNA-disease associations from the integrated circRNA similarity network and integrated disease similarity network, respectively. Fivefold cross-validation of four datasets shows that RNMFLP is superior to the state-of-the-art methods. In addition, case studies on lung cancer, hepatocellular carcinoma and colorectal cancer further demonstrate the reliability of our method to discover disease-related circRNAs.

Identification of miRNA-disease associations via deep forest ensemble learning based on autoencoder.

Increasing evidences show that the occurrence of human complex diseases is closely related to microRNA (miRNA) variation and imbalance. For this reason, predicting disease-related miRNAs is essential for the diagnosis and treatment of complex human diseases. Although some current computational methods can effectively predict potential disease-related miRNAs, the accuracy of prediction should be further improved. In our study, a new computational method via deep forest ensemble learning based on autoencoder (DFELMDA) is proposed to predict miRNA-disease associations. Specifically, a new feature representation strategy is proposed to obtain different types of feature representations (from miRNA and disease) for each miRNA-disease association. Then, two types of low-dimensional feature representations are extracted by two deep autoencoders for predicting miRNA-disease associations. Finally, two prediction scores of the miRNA-disease associations are obtained by the deep random forest and combined to determine the final results. DFELMDA is compared with several classical methods on the The Human microRNA Disease Database (HMDD) dataset. Results reveal that the performance of this method is superior. The area under receiver operating characteristic curve (AUC) values obtained by DFELMDA through 5-fold and 10-fold cross-validation are 0.9552 and 0.9560, respectively. In addition, case studies on colon, breast and lung tumors of different disease types further demonstrate the excellent ability of DFELMDA to predict disease-associated miRNA-disease. Performance analysis shows that DFELMDA can be used as an effective computational tool for predicting miRNA-disease associations.

DAESTB: inferring associations of small molecule-miRNA via a scalable tree boosting model based on deep autoencoder.

MicroRNAs (miRNAs) are closely related to a variety of human diseases, not only regulating gene expression, but also having an important role in human life activities and being viable targets of small molecule drugs for disease treatment. Current computational techniques to predict the potential associations between small molecule and miRNA are not that accurate. Here, we proposed a new computational method based on a deep autoencoder and a scalable tree boosting model (DAESTB), to predict associations between small molecule and miRNA. First, we constructed a high-dimensional feature matrix by integrating small molecule-small molecule similarity, miRNA-miRNA similarity and known small molecule-miRNA associations. Second, we reduced feature dimensionality on the integrated matrix using a deep autoencoder to obtain the potential feature representation of each small molecule-miRNA pair. Finally, a scalable tree boosting model is used to predict small molecule and miRNA potential associations. The experiments on two datasets demonstrated the superiority of DAESTB over various state-of-the-art methods. DAESTB achieved the best AUC value. Furthermore, in three case studies, a large number of predicted associations by DAESTB are confirmed with the public accessed literature. We envision that DAESTB could serve as a useful biological model for predicting potential small molecule-miRNA associations.