Search

In this study, we proposed an ensemble learning method, simultaneously integrating a low-rank matrix completion model and a ridge regression model to predict anticancer drug response on cancer cell lines. The model was applied to two benchmark datasets, including the Cancer Cell Line Encyclopedia (CCLE) and the Genomics of Drug Sensitivity in Cancer (GDSC). As previous studies suggest, the dual-layer integrated cell line-drug network model was one of the best models by far and outperformed most state-of-the-art models. Thus, we performed a head-to-head comparison between the dual-layer integrated cell line-drug network model and our model by a 10-fold crossvalidation study. For the CCLE dataset, our model has a higher Pearson correlation coefficient between predicted and observed drug responses than that of the dual-layer integrated cell line-drug network model in 18 out of 23 drugs. For the GDSC dataset, our model is better in 26 out of 28 drugs in the phosphatidylinositol 3-kinase (PI3K) pathway and 26 out of 30 drugs in the extracellular signal-regulated kinase (ERK) signaling pathway, respectively. Based on the prediction results, we carried out two types of case studies, which further verified the effectiveness of the proposed model on the drug-response prediction. In addition, our model is more biologically interpretable than the compared method, since it explicitly outputs the genes involved in the prediction, which are enriched in functions, like transcription, Src homology 2/3 (SH2/3) domain, cell cycle, ATP binding, and zinc finger., Graphical Abstract, This study proposed an ensemble learning method, simultaneously integrating a low-rank matrix completion model and a ridge regression model to predict anticancer drug response on cancer cell lines. The model was effectively applied to two benchmark datasets. It also explicitly outputs the genes involved in the prediction, which are enriched in functions.

This study aims to explore the moderating effect of the number of images and videos on the relationship between text length in crowdfunding campaign descriptions and crowdfunding outcomes. We use data from 13,622 technology campaigns on the Kickstarter website to test our hypotheses. The findings indicate that an increased number of images and videos improves fundraising performance but weakens the positive effect of description length on the fundraising performance of technology campaigns. In other words, on the Kickstarter website, we find an overshadowing effect between multimedia. The term “overshadowing effect” proposed in this paper suggests that redundant media can obscure the effect of other media in working memory. This study is among the first to explore the interaction between multimedia types from the theoretical perspective of cognitive load.

For resectable cancer patients, a method that could precisely predict the risk of postoperative recurrence would be crucial for guiding adjuvant treatment. Since T cell receptor (TCR) repertoires had been shown to be closely related to the dynamics of cancers, here we enrolled a cohort of patients to evaluate the potential of TCR repertoires in predicting the prognosis of resectable non-small cell lung cancers. Specifically, TCRβ repertoires were analyzed in surgical tumor tissues and matched adjacent non-tumor tissues from 39 patients enrolled with resectable non-small cell lung cancer, through target enrichment and high-throughput sequencing. As a result, there are significant differences between the TCR repertories of tumor samples and those of matched adjacent non-tumor samples as evaluated by criteria like the number of clonotypes. In addition, TCR repertoires were significantly associated with a few clinical features, as well as somatic mutations. Finally, certain TCRβ variable-joining (V-J) pairings were featured to build a logistic regression model in predicting postoperative recurrence of resectable non-small cell lung cancers with a testing area under the receiver operating characteristic curve (AUC) of around 0.9. Thus, we hypothesize that TCR repertoires could be potentially used to predict prognosis after curative surgery for non-small cell lung cancer patients., T cell receptor (TCR) repertoires have emerged as potential biomarkers in progress of multiple cancers. Chen and colleagues comprehensively characterized TCR repertoires in tumors from resectable non-small cell lung cancer (NSCLC) patients and evaluated the potential of TCR repertoires in predicting the prognosis of NSCLC who received curative surgery., Graphical Abstract

In recent years, more and more single-cell technologies have been developed. A vast amount of single-cell omics data has been generated by large projects, such as the Human Cell Atlas, the Mouse Cell Atlas, the Mouse RNA Atlas, the Mouse ATAC Atlas, and the Plant Cell Atlas. Based on these single-cell big data, thousands of bioinformatics algorithms for quality control, clustering, cell-type annotation, developmental inference, cell-cell transition, cell-cell interaction, and spatial analysis are developed. With powerful experimental single-cell technology and state-of-the-art big data analysis methods based on artificial intelligence, the molecular landscape at the single-cell level can be revealed. With spatial transcriptomics and single-cell multi-omics, even the spatial dynamic multi-level regulatory mechanisms can be deciphered. Such single-cell technologies have many successful applications in oncology, assisted reproduction, embryonic development, and plant breeding. We not only review the experimental and bioinformatics methods for single-cell research, but also discuss their applications in various fields and forecast the future directions for single-cell technologies. We believe that spatial transcriptomics and single-cell multi-omics will become the next booming business for mechanism research and commercial industry.

Multi-objective evolutionary algorithms based on decomposition (MOEA/D) decomposes a multi-objective optimization problem (MOP) into a set of simple scalar objective optimization sub-problems and solves them in a collaborative way. Since the sub-problems are different in optimization difficulty and computational resource demanding, it is critical to reasonably allocate computational resources among them, which can optimize the usage of resources and improve the performance of an algorithm. This paper proposes a new resource allocation strategy based on the relationship between sub-problems for MOEA/D. A probability vector is maintained based on the relationship between sub-problems, which is used to guide the selection of sub-problems for optimization. In the optimization process, we explored the role of priority optimization of boundary sub-problems and used it to assist in the update of probability vector in the early optimization phase. A steady-state algorithm is designed and tested experimentally. The results suggest that the designed algorithms have some advantages over existing state-of-the-art algorithms.

Subspace clustering, which is devoted to classifying data samples derived from a union of linear subspaces, has been widely applied to various fields such as pattern recognition, artificial intelligence and computer vision. In this paper, we propose a symmetry constrained latent low rank representation with converted nuclear norm (SLLRRC) algorithm for robust subspace clustering, which extends the original latent low rank representation (LLRR) algorithm by introducing a kind of converted nuclear norm and integrating strategy of the symmetric constraint. SLLRRC both enhances the sparsity of the coefficient matrix and guarantees weight consistency for each pair of data samples when seeking the low rank representation. The symmetric coefficient matrix that will no longer be overly dense is acquired by the inexact augmented Lagrange multipliers (IALM) method. With further exploiting the angular information of the principal directions of that, the sparse affinity matrix for spectral clustering is identified. Extensive experimental results on face clustering and motion segmentation datasets confirm the availability and robustness of the proposed algorithm, and it is highly competitive compared to the state-of-the-art subspace clustering algorithms.

Colorectal cancer (CRC), the 3rd most universal cancer globally, accounts for approximately 10% of newly diagnosed cancer incidences each year. Identifying biomarkers associated with CRC survival and predicting the survival of CRC patients are critical for personalized therapy. Existing studies on CRC survival are mainly based on single omics, studies using multi-omics to predict CRC survival are still vacant. To fill in this gap, we aim to identify biomarkers associated with CRC survival at mRNA, miRNA and tissue microbiome levels, and to evaluate the accuracy of potential biomarkers in predicting CRC survival.First, we collected 31 short-term survival (ST, less than 3 years) and 47 long-term survival (LT, longer than 3 years) CRC samples from the database, was named The Cancer Genome Atlas (TCGA). Then, we carried out bioinformatics analysis with collected multi-omics data: (1) comparing the bacterial community structures between ST and LT, (2) identifying differentially expressed mRNAs and miRNAs between ST and LT, and (3) exploring the relationship between bacteria and genes. Finally, we trained models based on multi-omics data to evaluate the performance of several omics data in predicting CRC survival.Among the compared omics data, microbiome of CRC tissue had the best predictive power on the three-year survival of CRC patients, the area under the receiver operating characteristic curve (AUC) is 0.755 with 10-fold Cross-Validation (CV). In addition, we screened out 26 differential microbial communities and 13 differential expression genes (DEGs) between ST and LT. Thermoanaerobacterium, Parabacteroides, Oceanicaulis, and Acetonema were more abundant in the ST, while Methylotenera, Candidatus_Riesia and Aquamicrobium were enriched in the LT. We also found that up-regulated genes were significantly enriched in ST group, but the down-regulated genes were enriched in the LT group.The tissue bacterial communities of CRC patients with different survival periods show significant differences, and the bacteria in tumour tissue of CRC are potential biomarkers for predicting the three-year survival of CRC patients.

About 30%-40% breast cancer patients suffer from recurrence and metastasis, even after targeted therapy like trastuzumab. Since breast cancer recurrence and metastasis are intrinsically related to mortality, it is critical to predict the recurrence and metastasis risk of an individual patient, which is essential for adjuvant therapy and early intervention. In this study, we developed a novel breast cancer recurrence and metastasis risk assessment framework from histopathological images using image features and machine learning technologies. The detection framework was applied on a manually collected clinical dataset from the Cancer Hospital, Chinese Academy of Medical Sciences, consisting of 127 breast cancer patients with known prognostic information; and further independently validated on 88 formalin-fixed, paraffin-embedded (FFPE) samples downloaded from The Cancer Genome Atlas (TCGA) with known recurrence and metastasis status. As a result, the XGBoost-based method performed well using only 8 texture and color features, obtained internal testing AUC of 0.75 on clinical data and external testing AUC of 0.72 on TCGA FFPE data, respectively. In addition, this study found two important potential predictors, i.e., the second moment of the B color component and the detail level mean square error of the wavelet multi-sub-bands co-occurrence matrix. Our study benchmarked the performances of histopathological image features and machine learning technologies in the recurrence and metastasis risk assessment, and holds promise for relieving pathologists' workload and boosting the survival chances of the breast cancer patients.

It is urgent to find an effective antiviral drug against SARS-CoV-2. In this study, 96 virus-drug associations (VDAs) from 12 viruses including SARS-CoV-2 and similar viruses and 78 small molecules are selected. Complete genomic sequence similarity of viruses and chemical structure similarity of drugs are then computed. A KATZ-based VDA prediction method (VDA-KATZ) is developed to infer possible drugs associated with SARS-CoV-2. VDA-KATZ obtained the best AUCs of 0.8803 when the walking length is 2. The predicted top 3 antiviral drugs against SARS-CoV-2 are remdesivir, oseltamivir, and zanamivir. Molecular docking is conducted between the predicted top 10 drugs and the virus spike protein/human ACE2. The results showed that the above 3 chemical agents have higher molecular binding energies with ACE2. For the first time, we found that zidovudine may be effective clues of treatment of COVID-19. We hope that our predicted drugs could help to prevent the spreading of COVID., Highlights • 96 virus-drug associations (VDAs) from 12 viruses including SARS-CoV-2 and similar viruses and 78 small molecules are selected. • Complete genomic sequence similarity of viruses and chemical structure similarity of drugs are computed. • A KATZ-based VDA prediction method (VDA-KATZ) is developed to infer possible antiviral drugs against SARS-CoV-2. • Remdesivir, oseltamivir, and zanamivir have higher molecular binding energies with ACE2.

Artificial intelligence (AI) coupled with promising machine learning (ML) techniques well known from computer science is broadly affecting many aspects of various fields including science and technology, industry, and even our day-to-day life. The ML techniques have been developed to analyze high-throughput data with a view to obtaining useful insights, categorizing, predicting, and making evidence-based decisions in novel ways, which will promote the growth of novel applications and fuel the sustainable booming of AI. This paper undertakes a comprehensive survey on the development and application of AI in different aspects of fundamental sciences, including information science, mathematics, medical science, materials science, geoscience, life science, physics, and chemistry. The challenges that each discipline of science meets, and the potentials of AI techniques to handle these challenges, are discussed in detail. Moreover, we shed light on new research trends entailing the integration of AI into each scientific discipline. The aim of this paper is to provide a broad research guideline on fundamental sciences with potential infusion of AI, to help motivate researchers to deeply understand the state-of-the-art applications of AI-based fundamental sciences, and thereby to help promote the continuous development of these fundamental sciences., Graphical abstract, Public summary • “Can machines think?” The goal of artificial intelligence (AI) is to enable machines to mimic human thoughts and behaviors, including learning, reasoning, predicting, and so on. • “Can AI do fundamental research?” AI coupled with machine learning techniques is impacting a wide range of fundamental sciences, including mathematics, medical science, physics, etc. • “How does AI accelerate fundamental research?” New research and applications are emerging rapidly with the support by AI infrastructure, including data storage, computing power, AI algorithms, and frameworks.

Can low-carbon technological innovation improve energy efficiency in China? Despite China’s attempts to improve energy efficiency through low-carbon technological innovation for several years, many scholars argue that innovation does not necessarily produce the expected outcomes. This study tests the heterogeneity of energy efficiency in China based on Data Envelopment Analysis (DEA) and machine learning and demonstrates the spatial spillover effect of regional agglomeration of low-carbon technology innovation on energy efficiency using the dynamic spatial Durbin model (DSDM). The results show that: (a) China's energy efficiency is heterogeneous in spatial distribution, and it has the characteristics of spilling over from the east to the central and western regions. (b) The direct effect of low-carbon technological innovation on energy efficiency is positive, but the indirect effect is negative, and the total effect is negative. The low-carbon technological innovation has an agglomeration effect, i.e., innovation resources tend to gather in higher energy efficiency regions. (c) Government influence, industrial structure, and foreign capital are key factors of low-carbon technology innovation. Based on these results, this study suggests the following policy recommendations: Narrow the low-carbon technology innovation gap between the developed and developing regions; promote the utilization efficiency of low-carbon technology innovation resources; open the low-carbon technological innovation to foreign investment and encourage low-carbon enterprises to cooperate with them.

Exergy model and the least exergy loss principle are introduced to analyze the exergy loss distribution in heat transfer and air/water flow process, for a combined server/cabinet cooling network. Based on the optimized case suggested by the exergy model, a multi-stage water cooling scheme is designed, with two-stage heat exchanger in server cabinet, and double-chiller in the water loop. Both thermal and energy performance of such cooling infrastructure has been tested in a data center in China, with the measured performance in good agreement with the designed values, and the exergy loss model has been validated.