### Recent Submissions

• #### Exploiting machine learning models to identify novel Alzheimer’s disease biomarkers and potential targets

(Scientific Reports, Springer Science and Business Media LLC, 2023-03-27) [Article]
We still do not have an effective treatment for Alzheimer's disease (AD) despite it being the most common cause of dementia and impaired cognitive function. Thus, research endeavors are directed toward identifying AD biomarkers and targets. In this regard, we designed a computational method that exploits multiple hub gene ranking methods and feature selection methods with machine learning and deep learning to identify biomarkers and targets. First, we used three AD gene expression datasets to identify 1/ hub genes based on six ranking algorithms (Degree, Maximum Neighborhood Component (MNC), Maximal Clique Centrality (MCC), Betweenness Centrality (BC), Closeness Centrality, and Stress Centrality), 2/ gene subsets based on two feature selection methods (LASSO and Ridge). Then, we developed machine learning and deep learning models to determine the gene subset that best distinguishes AD samples from the healthy controls. This work shows that feature selection methods achieve better prediction performances than the hub gene sets. Beyond this, the five genes identified by both feature selection methods (LASSO and Ridge algorithms) achieved an AUC = 0.979. We further show that 70% of the upregulated hub genes (among the 28 overlapping hub genes) are AD targets based on a literature review and six miRNA (hsa-mir-16-5p, hsa-mir-34a-5p, hsa-mir-1-3p, hsa-mir-26a-5p, hsa-mir-93-5p, hsa-mir-155-5p) and one transcription factor, JUN, are associated with the upregulated hub genes. Furthermore, since 2020, four of the six microRNA were also shown to be potential AD targets. To our knowledge, this is the first work showing that such a small number of genes can distinguish AD samples from healthy controls with high accuracy and that overlapping upregulated hub genes can narrow the search space for potential novel targets.
• #### Applications of Depth Minimization of Decision Trees Containing Hypotheses for Multiple-Value Decision Tables

(Entropy, MDPI AG, 2023-03-23) [Article]
In this research, we consider decision trees that incorporate standard queries with one feature per query as well as hypotheses consisting of all features’ values. These decision trees are used to represent knowledge and are comparable to those investigated in exact learning, in which membership queries and equivalence queries are used. As an application, we look into the issue of creating decision trees for two cases: the sorting of a sequence that contains equal elements and multiple-value decision tables which are modified from UCI Machine Learning Repository. We contrast the efficiency of several forms of optimal (considering the parameter depth) decision trees with hypotheses for the aforementioned applications. We also investigate the efficiency of decision trees built by dynamic programming and by an entropy-based greedy method. We discovered that the greedy algorithm produces very similar results compared to the results of dynamic programming algorithms. Therefore, since the dynamic programming algorithms take a long time, we may readily apply the greedy algorithms.
• #### 3D-Printed disposable nozzles for cost-efficient extrusion-based 3D bioprinting

(Materials Science in Additive Manufacturing, AccScience Publishing, 2023-03-21) [Article]
3D bioprinting has significantly impacted tissue engineering with its capability to create intricate structures with complex geometries that were difficult to replicate through traditional manufacturing techniques. Extrusion-based 3D bioprinting methods tend to be limited when creating complex structures using bioinks of low viscosity. However, the capacity for creating multi-material structures that have distinct properties could be unlocked through the mixture of two solutions before extrusion. This could be used to generate architectures with varying levels of stiffness and hydrophobicity, which could be utilized for regenerative medicine applications. Moreover, it allows for combining proteins and other biological materials in a single 3D-bioprinted structure. This paper presents a standardized fabrication method of disposable nozzle connectors (DNC) for 3D bioprinting with hydrogel-based materials. This method entails 3D printing connectors with dual inlets and a single outlet to mix the material internally. The connectors are compatible with conventional Luer lock needles, offering an efficient solution for nozzle replacement. IVZK (Ac-Ile-Val-Cha-Lys-NH2) peptide-based hydrogel materials were used as a bioink with the 3D-printed DNCs. Extrusion-based 3D bioprinting was employed to print shapes of varying complexities, demonstrating potential in achieving high print resolution, shape fidelity, and biocompatibility. Post-printing of human neonatal dermal fibroblasts, cell viability, proliferation, and metabolic activity were observed, which demonstrated the effectiveness of the proposed design and process for 3D bioprinting using low-viscosity bioinks.
• #### AB-Gen: Antibody Library Design with Generative Pre-trained Transformer and Deep Reinforcement Learning

(Cold Spring Harbor Laboratory, 2023-03-21) [Preprint]
Antibody leads must fulfill multiple desirable properties to be clinical candidates. Primarily due to the low throughput in the experimental procedure, the need for such multi-property optimization causes the bottleneck in preclinical antibody discovery and development, because addressing one issue usually causes another. We developed a reinforcement learning (RL) method, named AB-Gen, for antibody library design using a generative pre-trained Transformer (GPT) as the policy network of the RL agent. We showed that this model can learn the antibody space of heavy chain complementarity determining region 3 (CDRH3) and generate sequences with similar property distributions. Besides, when using HER2 as the target, the agent model of AB-Gen was able to generate novel CDRH3 sequences that fulfill multi-property constraints. 509 generated sequences were able to pass all property filters and three highly conserved residues were identified. The importance of these residues was further demonstrated by molecular dynamics simulations, which consolidated that the agent model was capable of grasping important information in this complex optimization task. Overall, the AB-Gen method is able to design novel antibody sequences with an improved success rate than the traditional propose-then-filter approach. It has the potential to be used in practical antibody design, thus empowering the antibody discovery and development process.
• #### A comprehensive benchmarking with practical guidelines for cellular deconvolution of spatial transcriptomics

(Nature Communications, Springer Science and Business Media LLC, 2023-03-21) [Article]
Spatial transcriptomics technologies are used to profile transcriptomes while preserving spatial information, which enables high-resolution characterization of transcriptional patterns and reconstruction of tissue architecture. Due to the existence of low-resolution spots in recent spatial transcriptomics technologies, uncovering cellular heterogeneity is crucial for disentangling the spatial patterns of cell types, and many related methods have been proposed. Here, we benchmark 18 existing methods resolving a cellular deconvolution task with 50 real-world and simulated datasets by evaluating the accuracy, robustness, and usability of the methods. We compare these methods comprehensively using different metrics, resolutions, spatial transcriptomics technologies, spot numbers, and gene numbers. In terms of performance, CARD, Cell2location, and Tangram are the best methods for conducting the cellular deconvolution task. To refine our comparative results, we provide decision-tree-style guidelines and recommendations for method selection and their additional features, which will help users easily choose the best method for fulfilling their concerns.
• #### A universal framework for single-cell multi-omics data integration with graph convolutional networks

(Briefings in bioinformatics, Oxford University Press (OUP), 2023-03-17) [Article]
Single-cell omics data are growing at an unprecedented rate, whereas effective integration of them remains challenging due to different sequencing methods, quality, and expression pattern of each omics data. In this study, we propose a universal framework for the integration of single-cell multi-omics data based on graph convolutional network (GCN-SC). Among the multiple single-cell data, GCN-SC usually selects one data with the largest number of cells as the reference and the rest as the query dataset. It utilizes mutual nearest neighbor algorithm to identify cell-pairs, which provide connections between cells both within and across the reference and query datasets. A GCN algorithm further takes the mixed graph constructed from these cell-pairs to adjust count matrices from the query datasets. Finally, dimension reduction is performed by using non-negative matrix factorization before visualization. By applying GCN-SC on six datasets, we show that GCN-SC can effectively integrate sequencing data from multiple single-cell sequencing technologies, species or different omics, which outperforms the state-of-the-art methods, including Seurat, LIGER, GLUER and Pamona.
• #### miProBERT: identification of microRNA promoters based on the pre-trained model BERT.

(Briefings in bioinformatics, Oxford University Press (OUP), 2023-03-17) [Article]
Accurate prediction of promoter regions driving miRNA gene expression has become a major challenge due to the lack of annotation information for pri-miRNA transcripts. This defect hinders our understanding of miRNA-mediated regulatory networks. Some algorithms have been designed during the past decade to detect miRNA promoters. However, these methods rely on biosignal data such as CpG islands and still need to be improved. Here, we propose miProBERT, a BERT-based model for predicting promoters directly from gene sequences without using any structural or biological signals. According to our information, it is the first time a BERT-based model has been employed to identify miRNA promoters. We use the pre-trained model DNABERT, fine-tune the pre-trained model on the gene promoter dataset so that the model includes information about the richer biological properties of promoter sequences in its representation, and then systematically scan the upstream regions of each intergenic miRNA using the fine-tuned model. About, 665 miRNA promoters are found. The innovative use of a random substitution strategy to construct a negative dataset improves the discriminative ability of the model and further reduces the false positive rate (FPR) to as low as 0.0421. On independent datasets, miProBERT outperformed other gene promoter prediction methods. With comparison on 33 experimentally validated miRNA promoter datasets, miProBERT significantly outperformed previously developed miRNA promoter prediction programs with 78.13% precision and 75.76% recall. We further verify the predicted promoter regions by analyzing conservation, CpG content and histone marks. The effectiveness and robustness of miProBERT are highlighted.
• #### An overlooked soil carbon pool in vegetated coastal ecosystems: National-scale assessment of soil organic carbon stocks in coastal shelter forests of China.

(The Science of the total environment, Elsevier BV, 2023-03-17) [Article]
Protection and restoration of vegetated coastal ecosystems provide opportunities to mitigate climate change. Coastal shelter forests as one of vegetated coastal ecosystems play vital role on sandy coasts protection, but less attention is paid on their soil organic carbon (OC) sequestration potential. Here, we provide the first national-scale assessment of the soil OC stocks, fractions, sources and accumulation rates from 48 sites of shelter forests and 74 sites of sandy beaches across 22° of latitude in China. We find that, compared with sandy beaches, shelter forest plantation achieves an average soil desalination rate of 92.0 % and reduces the soil pH by 1.3 units. The improved soil quality can facilitate OC sequestration leading to an increase of soil OC stock of 11.8 (0.60–64.2) MgC ha−1 in shelter forests. Particulate OC (POC) is a dominant OC fraction in both sandy beaches and shelter forests, but most sites are >80 % in shelter forests. The low δ13C values and higher C:N ratios, which are more regulated by climate and tree species, together with high POC proportions suggest a substantial contribution of plant-derived OC. Bayesian mixing model indicates that 71.8 (33.5–91.6)% of the soil OC is derived from local plant biomass. We estimate that soil OC stocks in Chinese shelter forests are 20.5 (7.44–79.7) MgC ha−1 and 4.53 ± 0.71 TgC in the top meter, with an accumulation rate of 45.0 (6.90 to 194.1) gC m−2 year−1 and 99.5 ± 44.9 GgC year−1. According to coastal shelter forest afforestation plan, additional 1.72 ± 0.27 TgC with a rate of 37.9 ± 17.1 GgC year−1 can be sequestrated in the future. Our findings suggest that construction of coastal shelter forests can be an effective solution to sequester more soil carbon in coastal ecosystems.
• #### Efficient in planta production of amidated antimicrobial peptides that are active against drug-resistant ESKAPE pathogens

(Nature Communications, Springer Science and Business Media LLC, 2023-03-16) [Article]
Antimicrobial peptides (AMPs) are promising next-generation antibiotics that can be used to combat drug-resistant pathogens. However, the high cost involved in AMP synthesis and their short plasma half-life render their clinical translation a challenge. To address these shortcomings, we report efficient production of bioactive amidated AMPs by transient expression of glycine-extended AMPs in Nicotiana benthamiana line expressing the mammalian enzyme peptidylglycine α-amidating mono-oxygenase (PAM). Cationic AMPs accumulate to substantial levels in PAM transgenic plants compare to nontransgenic N. benthamiana. Moreover, AMPs purified from plants exhibit robust killing activity against six highly virulent and antibiotic resistant ESKAPE pathogens, prevent their biofilm formation, analogous to their synthetic counterparts and synergize with antibiotics. We also perform a base case techno-economic analysis of our platform, demonstrating the potential economic advantages and scalability for industrial use. Taken together, our experimental data and techno-economic analysis demonstrate the potential use of plant chassis for large-scale production of clinical-grade AMPs.
• #### The Influence of Prenatal Exposure to Methamphetamine on the Development of Dopaminergic Neurons in the Ventral Midbrain

(International Journal of Molecular Sciences, MDPI AG, 2023-03-16) [Article]
Methamphetamine, a highly addictive central nervous system (CNS) stimulant, is used worldwide as an anorexiant and attention enhancer. Methamphetamine use during pregnancy, even at therapeutic doses, may harm fetal development. Here, we examined whether exposure to methamphetamine affects the morphogenesis and diversity of ventral midbrain dopaminergic neurons (VMDNs). The effects of methamphetamine on morphogenesis, viability, the release of mediator chemicals (such as ATP), and the expression of genes involved in neurogenesis were evaluated using VMDNs isolated from the embryos of timed-mated mice on embryonic day 12.5. We demonstrated that methamphetamine (10 µM; equivalent to its therapeutic dose) did not affect the viability and morphogenesis of VMDNs, but it reduced the ATP release negligibly. It significantly downregulated Lmx1a, En1, Pitx3, Th, Chl1, Dat, and Drd1 but did not affect Nurr1 or Bdnf expression. Our results illustrate that methamphetamine could impair VMDN differentiation by altering the expression of important neurogenesis-related genes. Overall, this study suggests that methamphetamine use may impair VMDNs in the fetus if taken during pregnancy. Therefore, it is essential to exercise strict caution for its use in expectant mothers.
• #### Molecular insights into the Darwin paradox of coral reefs from the sea anemone Aiptasia

(Science Advances, American Association for the Advancement of Science (AAAS), 2023-03-15) [Article]
Symbiotic cnidarians such as corals and anemones form highly productive and biodiverse coral reef ecosystems in nutrient-poor ocean environments, a phenomenon known as Darwin’s paradox. Resolving this paradox requires elucidating the molecular bases of efficient nutrient distribution and recycling in the cnidarian-dinoflagellate symbiosis. Using the sea anemone Aiptasia, we show that during symbiosis, the increased availability of glucose and the presence of the algae jointly induce the coordinated up-regulation and relocalization of glucose and ammonium transporters. These molecular responses are critical to support symbiont functioning and organism-wide nitrogen assimilation through glutamine synthetase/glutamate synthase–mediated amino acid biosynthesis. Our results reveal crucial aspects of the molecular mechanisms underlying nitrogen conservation and recycling in these organisms that allow them to thrive in the nitrogen-poor ocean environments.
• #### LEP-AD: Language Embedding of Proteins and Attention to Drugs predicts drug target interactions

(Cold Spring Harbor Laboratory, 2023-03-15) [Preprint]
Predicting drug-target interactions is a tremendous challenge for drug development and lead optimization. Recent advances include training algorithms to learn drug-target interactions from data and molecular simulations. Here we utilize Evolutionary Scale Modeling (ESM-2) models to establish a Transformer protein language model for drug-target interaction predictions. Our architecture, LEP- AD, combines pre-trained ESM-2 and Transformer-GCN models predicting bind-ing affinity values. We report new best-in-class state-of-the-art results compared to competing methods such as SimBoost, DeepCPI, Attention-DTA, GraphDTA, and more using multiple datasets, including Davis, KIBA, DTC, Metz, ToxCast, and STITCH. Finally, we find that a pre-trained model with embedding of proteins (the LED-AD) outperforms a model using an explicit alpha-fold 3D representation of proteins (e.g., LEP-AD supervised by Alphafold). The LEP-AD model scales favorably in performance with the size of training data.
• #### Diabetic cardiomyopathy: The role of microRNAs and long non-coding RNAs

(Frontiers in Endocrinology, Frontiers Media SA, 2023-03-07) [Article]
Diabetes mellitus (DM) is on the rise, necessitating the development of novel therapeutic and preventive strategies to mitigate the disease’s debilitating effects. Diabetic cardiomyopathy (DCMP) is among the leading causes of morbidity and mortality in diabetic patients globally. DCMP manifests as cardiomyocyte hypertrophy, apoptosis, and myocardial interstitial fibrosis before progressing to heart failure. Evidence suggests that non-coding RNAs, such as long non-coding RNAs (lncRNAs) and microRNAs (miRNAs), regulate diabetic cardiomyopathy-related processes such as insulin resistance, cardiomyocyte apoptosis and inflammation, emphasizing their heart-protective effects. This paper reviewed the literature data from animal and human studies on the non-trivial roles of miRNAs and lncRNAs in the context of DCMP in diabetes and demonstrated their future potential in DCMP treatment in diabetic patients.
• #### Intelligent and smart biomaterials for sustainable 3D printing applications

(Current Opinion in Biomedical Engineering, Elsevier BV, 2023-03-04) [Article]
Smart and intelligent biomaterials can be designed to carry out special tasks in modern medicine and sustainability, and engineered to identify and respond to environmental stimuli. Therefore, intelligent biomaterials have a large number of applications that can go from health (e.g. tissue engineering, drug delivery and biosensors), to more recently explored environmental applications involving ecosystem restoration (e.g. coral reefs and environmental remediation). The use of 3D printing technology opens the vision towards automated biomanufacturing with more precision and definition. With this broad range of applications, smart and intelligent biomaterials are used separately or in combination with 3D printing to enable the design of eco-friendly and sustainable solutions that can be used to overcome challenges for both; modern medicine and the environment.
• #### A learning-based image processing approach for pulse wave velocity estimation using spectrogram from peripheral pulse wave signals: An in silico study

(Frontiers in physiology, Frontiers Media SA, 2023-03-03) [Article]
Carotid-to-femoral pulse wave velocity (cf-PWV) is considered a critical index to evaluate arterial stiffness. For this reason, estimating Carotid-to-femoral pulse wave velocity (cf-PWV) is essential for diagnosing and analyzing different cardiovascular diseases. Despite its broader adoption in the clinical routine, the measurement process of carotid-to-femoral pulse wave velocity is considered a demanding task for clinicians and patients making it prone to inaccuracies and errors in the estimation. A smart non-invasive, and peripheral measurement of carotid-to-femoral pulse wave velocity could overcome the challenges of the classical assessment process and improve the quality of patient care. This paper proposes a novel methodology for the carotid-to-femoral pulse wave velocity estimation based on the use of the spectrogram representation from single non-invasive peripheral pulse wave signals [photoplethysmography (PPG) or blood pressure (BP)]. This methodology was tested using three feature extraction methods based on the semi-classical signal analysis (SCSA) method, the Law’s mask for texture energy extraction, and the central statistical moments. Finally, each feature method was fed into different machine learning models for the carotid-to-femoral pulse wave velocity estimation. The proposed methodology obtained an $R2\geq0.90$ for all the peripheral signals for the noise-free case using the MLP model, and for the different noise levels added to the original signal, the SCSA-based features with the MLP model presented an $R2\geq0.91$ for all the peripheral signals at the level of noise. These results provide evidence of the capacity of spectrogram representation for efficiently assessing the carotid-to-femoral pulse wave velocity estimation using different feature methods. Future work will be done toward testing the proposed methodology for in-vivo signals.
• #### Computational network analysis of host genetic risk variants of severe COVID-19.

(Human genomics, Springer Science and Business Media LLC, 2023-03-02) [Article]
Background: Genome-wide association studies have identified numerous human host genetic risk variants that play a substantial role in the host immune response to SARS-CoV-2. Although these genetic risk variants significantly increase the severity of COVID-19, their influence on body systems is poorly understood. Therefore, we aim to interpret the biological mechanisms and pathways associated with the genetic risk factors and immune responses in severe COVID-19. We perform a deep analysis of previously identified risk variants and infer the hidden interactions between their molecular networks through disease mapping and the similarity of the molecular functions between constructed networks. Results: We designed a four-stage computational workflow for systematic genetic analysis of the risk variants. We integrated the molecular profiles of the risk factors with associated diseases, then constructed protein–protein interaction networks. We identified 24 protein–protein interaction networks with 939 interactions derived from 109 filtered risk variants in 60 risk genes and 56 proteins. The majority of molecular functions, interactions and pathways are involved in immune responses; several interactions and pathways are related to the metabolic and cardiovascular systems, which could lead to multi-organ complications and dysfunction. Conclusions: This study highlights the importance of analyzing molecular interactions and pathways to understand the heterogeneous susceptibility of the host immune response to SARS-CoV-2. We propose new insights into pathogenicity analysis of infections by including genetic risk information as essential factors to predict future complications during and after infection. This approach may assist more precise clinical decisions and accurate treatment plans to reduce COVID-19 complications.
• #### A thermophilic chemolithoautotrophic bacterial consortium suggests a mutual relationship between bacteria in extreme oligotrophic environments

(Communications Biology, Springer Science and Business Media LLC, 2023-03-01) [Article]
A thermophilic, chemolithoautotrophic, and aerobic microbial consortium (termed carbonitroflex) growing in a nutrient-poor medium and an atmosphere containing N2, O2, CO2, and CO is investigated as a model to expand our understanding of extreme biological systems. Here we show that the consortium is dominated by Carbonactinospora thermoautotrophica (strain StC), followed by Sphaerobacter thermophilus, Chelatococcus spp., and Geobacillus spp. Metagenomic analysis of the consortium reveals a mutual relationship among bacteria, with C. thermoautotrophica StC exhibiting carboxydotrophy and carbon-dioxide storage capacity. C. thermoautotrophica StC, Chelatococcus spp., and S. thermophilus harbor genes encoding CO dehydrogenase and formate oxidase. No pure cultures were obtained under the original growth conditions, indicating that a tightly regulated interactive metabolism might be required for group survival and growth in this extreme oligotrophic system. The breadwinner hypothesis is proposed to explain the metabolic flux model and highlight the vital role of C. thermoautotrophica StC (the sole keystone species and primary carbon producer) in the survival of all consortium members. Our data may contribute to the investigation of complex interactions in extreme environments, exemplifying the interconnections and dependency within microbial communities.
• #### Genetic Variants in Protein Tyrosine Phosphatase Non-Receptor Type 23 Are Responsible for Mesiodens Formation

(Biology, MDPI AG, 2023-03-01) [Article]
A mesiodens is a supernumerary tooth located in the midline of the premaxilla. In order to investigate the genetic etiology of mesiodens, clinical and radiographic examination and whole exome sequencing (WES) were performed in 24 family members of a two-generation Hmong family and additionally in two unrelated Thai patients with mesiodens. WES in the Hmong family revealed a missense mutation (c.1807G>A;p.Glu603Lys) in PTPN23 in seven affected members and six unaffected members. The mode of inheritance was autosomal dominance with incomplete penetrance (53.84%). Two additional mutations in PTPN23, c.2248C>G;p.Pro750Ala and c.3298C>T;p.Arg1100Cys were identified in two unrelated patients with mesiodens. PTPN23 is a regulator of endosomal trafficking functioning to move activated membrane receptors, such as EGFR, from the endosomal sorting complex towards the ESCRT-III complex for multivesicular body biogenesis, lysosomal degradation, and subsequent downregulation of receptor signaling. Immunohistochemical study and RNAscope on developing mouse embryos showed broad expression of PTPN23 in oral tissues, while immunofluorescence showed that EGFR was specifically concentrated in the midline epithelium. Importantly, PTPN23 mutant protein was shown to have reduced phosphatase activity. In conclusion, mesiodens were associated with genetic variants in PTPN23, suggesting that mesiodens may form due to defects in endosomal trafficking, leading to disrupted midline signaling.
• #### Mycobiome structure does not affect field litter decomposition in Eucalyptus and Acacia plantations

(Frontiers in microbiology, Frontiers Media SA, 2023-02-28) [Article]
Mixed tree plantations have been studied because of their potential to improve biomass production, ecosystem diversity, and soil quality. One example is a mixture of Eucalyptus and Acacia trees, which is a promising strategy to improve microbial diversity and nutrient cycling in soil. We examined how a mixture of these species may influence the biochemical attributes and fungal community associated with leaf litter, and the effects on litter decomposition. We studied the litter from pure and mixed plantations, evaluating the effects of plant material and incubation site on the mycobiome and decomposition rate using litterbags incubated in situ. Our central hypothesis was litter fungal community would change according to incubation site, and it would interfere in litter decomposition rate. Both the plant material and the incubation locale significantly affected the litter decomposition. The origin of the litter was the main modulator of the mycobiome, with distinct communities from one plant species to another. The community changed with the incubation time but the incubation site did not influence the mycobiome community. Our data showed that litter and soil did not share the main elements of the community. Contrary to our hypothesis, the microbial community structure and diversity lacked any association with the decomposition rate. The differences in the decomposition pattern are explained basically as a function of the exchange of nitrogen compounds between the litter.
• #### Telecommunication Traffic Forecasting via Multi-task Learning

(ACM, 2023-02-27) [Conference Paper]
Accurate telecommunication time series forecasting is critical for smart management systems of cellular networks, and has a special challenge in predicting different types of time series simultaneously at one base station (BS), e.g., the SMS, Calls, and Internet. Unlike the well-studied single target forecasting problem for one BS, this distributed multi-target forecasting problem should take advantage of both the intra-BS dependence of different types of time series at the same BS and the inter-BS dependence of time series at different BS. To this end, we first propose a model to learn the inter-BS dependence by aggregating the multi-view dependence, e.g., from the viewpoint of SMS, Calls, and Internet. To incorporate the interBS dependence in time series forecasting, we then propose a Graph Gate LSTM (GGLSTM) model that includes a graph-based gate mechanism to unite those base stations with a strong dependence on learning a collaboratively strengthened prediction model. We also extract the intra-BS dependence by an attention network and use it in the final prediction. Our proposed approach is evaluated on two real-world datasets. Experiment results demonstrate the effectiveness of our model in predicting multiple types of telecom traffic at the distributed base stations.