Major histocompatibility complex (MHC) molecules are expressed on the cell surface, where they present peptides to T cells, which gives them a key role in the development of T-cell immune responses. MHC molecules come in two main variants: MHC Class I (MHC-I) and MHC Class II (MHC-II). MHC-I predominantly present peptides derived from intracellular proteins, whereas MHC-II predominantly presents peptides from extracellular proteins. In both cases, the binding between MHC and antigenic peptides is the most selective step in the antigen presentation pathway. Therefore, the prediction of peptide binding to MHC is a powerful utility to predict the possible specificity of a T-cell immune response. Commonly MHC binding prediction tools are trained on binding affinity or mass spectrometry-eluted ligands. Recent studies have however demonstrated how the integration of both data types can boost predictive performances. Inspired by this, we here present NetMHCpan-4.1 and NetMHCIIpan-4.0, two web servers created to predict binding between peptides and MHC-I and MHC-II, respectively. Both methods exploit tailored machine learning strategies to integrate different training data types, resulting in state-of-the-art performance and outperforming their competitors. The servers are available at http://www.cbs.dtu.dk/services/NetMHCpan-4.1/ and http://www.cbs.dtu.dk/services/NetMHCIIpan-4.0/. <<<

We introduce a new generative model where samples are produced via Langevin dynamics using gradients of the data distribution estimated with score matching. Because gradients can be ill-defined and hard to estimate when the data resides on low-dimensional manifolds, we perturb the data with different levels of Gaussian noise, and jointly estimate the corresponding scores, i.e., the vector fields of gradients of the perturbed data distribution for all noise levels. For sampling, we propose an annealed Langevin dynamics where we use gradients corresponding to gradually decreasing noise levels as the sampling process gets closer to the data manifold. Our framework allows flexible model architectures, requires no sampling during training or the use of adversarial methods, and provides a learning objective that can be used for principled model comparisons. Our models produce samples comparable to GANs on MNIST, CelebA and CIFAR-10 datasets, achieving a new state-of-the-art inception score of 8.87 on CIFAR-10. Additionally, we demonstrate that our models learn effective representations via image inpainting experiments. <<<

Denise M Wolf, Christina Yau, Julia Wulfkuhle, Lamorna Brown-Swigart, Rosa I Gallagher, Pei Rong Evelyn Lee, Zelos Zhu, Mark J Magbanua, Rosalyn Sayaman, Nicholas O'Grady, Amrita Basu, Amy Delson, Jean Philippe Coppé, Ruixiao Lu, Jerome Braun, I-SPY2 Investigators, Smita M Asare, Laura Sit, Jeffrey B Matthews, Jane Perlmutter, Nola Hylton, Minetta C Liu, Paula Pohlmann, W Fraser Symmans, Hope S Rugo, Claudine Isaacs, Angela M DeMichele, Douglas Yee, Donald A Berry, Lajos Pusztai, Emanuel F Petricoin, Gillian L Hirst, Laura J Esserman, Laura J van 't Veer <<<

Using pre-treatment gene expression, protein/phosphoprotein, and clinical data from the I-SPY2 neoadjuvant platform trial (NCT01042379), we create alternative breast cancer subtypes incorporating tumor biology beyond clinical hormone receptor (HR) and human epidermal growth factor receptor-2 (HER2) status to better predict drug responses. We assess the predictive performance of mechanism-of-action biomarkers from ∼990 patients treated with 10 regimens targeting diverse biology. We explore >11 subtyping schemas and identify treatment-subtype pairs maximizing the pathologic complete response (pCR) rate over the population. The best performing schemas incorporate Immune, DNA repair, and HER2/Luminal phenotypes. Subsequent treatment allocation increases the overall pCR rate to 63% from 51% using HR/HER2-based treatment selection. pCR gains from reclassification and improved patient selection are highest in HR subsets (>15%). As new treatments are introduced, the subtyping schema determines the minimum response needed to show efficacy. This data platform provides an unprecedented resource and supports the usage of response-based subtypes to guide future treatment prioritization. <<<

In recent years, scientists have increasingly taken to investigate the predictive nature of cognition. We argue that prediction relies on abstraction, and thus theories of predictive cognition need an explicit theory of abstract representation. We propose such a theory of the abstract representational capacities that allow humans to transcend the "here-and-now." Consistent with the predictive cognition literature, we suggest that the representational substrates of the mind are built as a hierarchy, ranging from the concrete to the abstract; however, we argue that there are qualitative differences between elements along this hierarchy, generating meaningful, often unacknowledged, diversity. Echoing views from philosophy, we suggest that the representational hierarchy can be parsed into: modality-specific representations, instantiated on perceptual similarity; multimodal representations, instantiated primarily on the discovery of spatiotemporal contiguity; and categorical representations, instantiated primarily on social interaction. These elements serve as the building blocks of complex structures discussed in cognitive psychology (e.g., episodes, scripts) and are the inputs for mental representations that behave like functions, typically discussed in linguistics (i.e., predicators). We support our argument for representational diversity by explaining how the elements in our ontology are all required to account for humans' predictive cognition (e.g., in subserving logic-based prediction; in optimizing the trade-off between accurate and detailed predictions) and by examining how the neuroscientific evidence coheres with our account. In doing so, we provide a testable model of the neural bases of conceptual cognition and highlight several important implications to research on self-projection, reinforcement learning, and predictive-processing models of psychopathology. <<<

Bioactivities, such as freshness maintenance, whitening, and prebiotics, of marine neoagaro-oligosaccharides (NAOS) with 4-12 degrees of polymerization (DPs) have been proven. However, NAOS produced by most marine β-agarases always possess low DPs (≤6) and limited categories; thus, a strategy that can efficiently produce NAOS especially with various DPs ≥8 must be developed. In this study, 60 amino acid residues with no functional annotation result were removed from the C-terminal of agarase AgaM1, and truncated recombinant AgaM1 (trAgaM1) was found to have the ability to produce NAOS with various DPs (4-12) under certain conditions. The catalytic efficiency and stability of trAgaM1 were obviously lower than the wild type (rAgaM1), which probably endowed trAgaM1 with the ability to produce NAOS with various DPs. The optimum conditions for various NAOS production included mixing 1% agarose (w/v) with 10.26 U/ml trAgaM1 and incubating the mixture at 50°C in deionized water for 100 min. This strategy produced neoagarotetraose (NA4), neoagarohexaose (NA6), neoagarooctaose (NA8), neoagarodecaose (NA10), and neoagarododecaose (NA12) at final concentrations of 0.15, 1.53, 1.53, 3.02, and 3.02 g/L, respectively. The NAOS served as end-products of the reaction. The conditions for trAgaM1 expression in a shake flask and 5 L fermentation tank were optimized, and the yields of trAgaM1 increased by 56- and 842-fold compared with those before optimization, respectively. This study provides numerous substrate sources for production and activity tests of NAOS with high DPs and offers a foundation for large-scale production of NAOS with various DPs at a low cost. <<<

AID/APOBEC family enzymes are best known for deaminating cytosine bases to uracil in single-stranded DNA, with characteristic sequence preferences that can produce mutational signatures in targets such as retroviral and cancer cell genomes. These deaminases have also been proposed to function in DNA demethylation via deamination of either 5-methylcytosine (mC) or TET-oxidized mC bases (ox-mCs), which include 5-hydroxymethylcytosine, 5-formylcytosine and 5-carboxylcytosine. One specific family member, APOBEC3A (A3A), has been shown to readily deaminate mC, raising the prospect of broader activity on ox-mCs. To investigate this claim, we developed a novel assay that allows for parallel profiling of activity on all modified cytosines. Our steady-state kinetic analysis reveals that A3A discriminates against all ox-mCs by >3700-fold, arguing that ox-mC deamination does not contribute substantially to demethylation. A3A is, by contrast, highly proficient at C/mC deamination. Under conditions of excess enzyme, C/mC bases can be deaminated to completion in long DNA segments, regardless of sequence context. Interestingly, under limiting A3A, the sequence preferences observed with targeting unmodified cytosine are further exaggerated when deaminating mC. Our study informs how methylation, oxidation, and deamination can interplay in the genome and suggests A3A's potential utility as a biotechnological tool to discriminate between cytosine modification states. <<<

BACKGROUND: Artesunate (ART) has been used extensively as anti-malarial drugs worldwide. Besides, it has also been reported to have anti-cancer activities. This study was aimed to explore the anti-cancer activity of ART in combination with cisplatin (CIS) on A549 cells.METHODS: Cells were cultured with different concentrations of ART and/or CIS for 24, 48, or 72 h to test the anti-proliferative effects by CCK-8 assay. Colony formation assay and EdU staining were also performed. TUNEL staining was used to illustrate the morphologic changes. Cell cycle and apoptosis were determined by flow cytometry assay, and Western blot analysis was conducted to detect the expression of apoptosis- and proliferation-related proteins. Caspase activities were determined by colorimetric assay kit. Moreover, the synergistic effect of ART with CIS in A549 cell xenograft model was also determined.RESULTS: ART significantly inhibited cell proliferation in dose- and time-dependent manners. Collectively, the combination treatment remarkably decreased colony formation rates and increased the rates of TUNEL-positive cells compared with mono-treatment. Mechanistically, the combination treatment influenced the expression of Bcl-2, Bax, p-P53, Caspase-3/7, Caspase-9, CyclinB1, P34, P21, and synergistically regulated the activity of P38/JNK/ERK1/2 MAPK pathway. In mice A549 xenograft tumors, the combination strategy significantly increased the anti-cancer efficacy of ART and CIS alone, consistent with the in vitro observations.CONCLUSIONS: ART exhibited significant anti-tumor effect on A549 cells and this efficiency could be enhanced by combination with CIS. <<<

基于物理信息的神经网络(Physics-informed Neural Networks,PINN),是一类用于解决有监督学习任务的神经网络,它不仅尽力遵循训练数据样本的分布规律,而且遵守由偏微分方程描述的物理定律。与纯数据驱动的神经网络学习相比,PINN在训练过程中施加了物理信息约束,因此能用更少的数据样本学习得到更具泛化能力的模型。近年来,PINN已逐渐成为机器学习和计算数学交叉学科的研究热点,并在理论和应用方面都获得了相对深入的研究,取得了可观的进展。但PINN独特的网络结构在实际应用中也存在训练缓慢甚至不收敛、精度低等问题。文中在总结当前PINN研究的基础上,对其网络/体系设计及其在流体力学等多个领域中的应用进行了探究,并展望了进一步的研究方向。 <<<

In the context of science, the well-known adage "a picture is worth a thousand words" might well be "a model is worth a thousand datasets." In this manuscript we introduce the SciML software ecosystem as a tool for mixing the information of physical laws and scientific models with data-driven machine learning approaches. We describe a mathematical object, which we denote universal differential equations (UDEs), as the unifying framework connecting the ecosystem. We show how a wide variety of applications, from automatically discovering biological mechanisms to solving high-dimensional Hamilton-Jacobi-Bellman equations, can be phrased and efficiently handled through the UDE formalism and its tooling. We demonstrate the generality of the software tooling to handle stochasticity, delays, and implicit constraints. This funnels the wide variety of SciML applications into a core set of training mechanisms which are highly optimized, stabilized for stiff equations, and compatible with distributed parallelism and GPU accelerators. <<<

In this work we investigate the use of the Signature Transform in the context of Learning. Under this assumption, we advance a supervised framework that potentially provides state-of-the-art classification accuracy with the use of few labels without the need of credit assignment and with minimal or no overfitting. We leverage tools from harmonic analysis by the use of the signature and log-signature, and use as a score function RMSE and MAE Signature and log-signature. We develop a closed-form equation to compute probably good optimal scale factors, as well as the formulation to obtain them by optimization. Techniques of Signal Processing are addressed to further characterize the problem. Classification is performed at the CPU level orders of magnitude faster than other methods. We report results on AFHQ, MNIST and CIFAR10, achieving 100% accuracy on all tasks assuming we can determine at test time which probably good optimal scale factor to use for each category. <<<

BACKGROUND: Tumor response to therapy is affected by both the cell types and the cell states present in the tumor microenvironment. This is true for many cancer treatments, including immune checkpoint inhibitors (ICIs). While it is well-established that ICIs promote T cell activation, their broader impact on other intratumoral immune cells is unclear; this information is needed to identify new mechanisms of action and improve ICI efficacy. Many preclinical studies have begun using single-cell analysis to delineate therapeutic responses in individual immune cell types within tumors. One major limitation to this approach is that therapeutic mechanisms identified in preclinical models have failed to fully translate to human disease, restraining efforts to improve ICI efficacy in translational research.METHOD: We previously developed a computational transfer learning approach called projectR to identify shared biology between independent high-throughput single-cell RNA-sequencing (scRNA-seq) datasets. In the present study, we test this algorithm's ability to identify conserved and clinically relevant transcriptional changes in complex tumor scRNA-seq data and expand its application to the comparison of scRNA-seq datasets with additional data types such as bulk RNA-seq and mass cytometry.RESULTS: We found a conserved signature of NK cell activation in anti-CTLA-4 responsive mouse and human tumors. In human metastatic melanoma, we found that the NK cell activation signature associates with longer overall survival and is predictive of anti-CTLA-4 (ipilimumab) response. Additional molecular approaches to confirm the computational findings demonstrated that human NK cells express CTLA-4 and bind anti-CTLA-4 antibodies independent of the antibody binding receptor (FcR) and that similar to T cells, CTLA-4 expression by NK cells is modified by cytokine-mediated and target cell-mediated NK cell activation.CONCLUSIONS: These data demonstrate a novel application of our transfer learning approach, which was able to identify cell state transitions conserved in preclinical models and human tumors. This approach can be adapted to explore many questions in cancer therapeutics, enhance translational research, and enable better understanding and treatment of disease. <<<

随着人均预期寿命的延长和受教育年限的提高,包括中国在内的世界各国劳动者的实际退休年龄却普遍低于法定退休年龄,呈现提前退休趋势。本文试图在生育受到约束的制度环境下对上述反常现象进行理论解释,通过构建一个动态一般均衡世代交叠(OLG)模型,考察预期寿命延长对劳动者的人力资本投资、退休年龄选择的影响机制,并结合中国的现实经济进行数值模拟。研究表明:当预期寿命提高时,人力资本投资能够在生命周期中获得更高的工资率回报,劳动者在少年期倾向于进行更多的人力资本投资,提高受教育年限;人力资本积累、有效工资率上升和利率下降引起的收入效应超过了替代效应,使得劳动者在老年期增加对闲暇的需求,有能力和意愿提早退休,减少终生劳动供给时间;进一步地,本文发现放松生育控制政策也会使劳动者享受更多闲暇时间的意愿增强,选择提前退休。本文还考察了个体退休行为的异质性,发现随着预期寿命的延长,相对穷人而言,富人的退休年龄和终生劳动供给更低。本文的结论有益于厘清劳动者在老年期的退休决策机理,为如何推进退休制度改革提供理论依据。 <<<

Wu Xiong, Alexandre Jousset, Rong Li, Manuel Delgado-Baquerizo, Mohammad Bahram, Ramiro Logares, Benjamin Wilden, Gerard Arjen de Groot, Nathalie Amacker, George A Kowalchuk, Qirong Shen, Stefan Geisen <<<

The colossal project of mapping the microbiome on Earth is rapidly advancing, with a focus on individual microbial groups. However, a global assessment of the associations between predatory protists and their bacterial prey is still missing at a cross-ecosystem level. This knowledge is critical to better understand the importance of top-down links in structuring microbiomes. Here, we examined 38 sequence-based datasets of paired bacterial and protistan taxa, covering 3,178 samples from diverse habitats including freshwater, marine and soils. We show that community profiles of protists and bacteria strongly correlated across and within habitats, with trophic microbiome structures fundamentally differing across habitats. Soils hosted the most heterogenous and diverse microbiomes. Protist communities were dominated by predators in soils and phototrophs in aquatic environments. This led to changes in the ratio of total protists to bacteria richness, which was highest in marine, while that of predatory protists to bacteria was highest in soils. Taxon richness and relative abundance of predatory protists positively correlated with bacterial richness in marine habitats. These links differed between soils, predatory protist richness and the relative abundance of predatory protists positively correlated with bacterial richness in forest and grassland soils, but not in agricultural soils. Our results suggested that anthropogenic pressure affects higher trophic levels more than lower ones leading to a decoupled trophic structure in microbiomes. Together, our cumulative overview of microbiome patterns of bacteria and protists at the global scale revealed major patterns and differences of the trophic structure of microbiomes across Earth's habitats, and show that anthropogenic factors might have negative effects on the trophic structure within microbiomes. Furthermore, the increased impact of anthropogenic factors on especially higher trophic levels suggests that often-observed reduced ecosystem functions in anthropogenic systems might be partly attributed to a reduction of trophic complexity. <<<

In coming to understand the world-in learning concepts, acquiring language, and grasping causal relations-our minds make inferences that appear to go far beyond the data available. How do we do it? This review describes recent approaches to reverse-engineering human learning and cognitive development and, in parallel, engineering more humanlike machine learning systems. Computational models that perform probabilistic inference over hierarchies of flexibly structured representations can address some of the deepest questions about the nature and origins of human thought: How does abstract knowledge guide learning and reasoning from sparse data? What forms does our knowledge take, across different domains and tasks? And how is that abstract knowledge itself acquired? <<<

V Gopalakrishnan, C N Spencer, L Nezi, A Reuben, M C Andrews, T V Karpinets, P A Prieto, D Vicente, K Hoffman, S C Wei, A P Cogdill, L Zhao, C W Hudgens, D S Hutchinson, T Manzo, M Petaccia de Macedo, T Cotechini, T Kumar, W S Chen, S M Reddy, R Szczepaniak Sloane, J Galloway-Pena, H Jiang, P L Chen, E J Shpall, K Rezvani, A M Alousi, R F Chemaly, S Shelburne, L M Vence, P C Okhuysen, V B Jensen, A G Swennes, F McAllister, E Marcelo Riquelme Sanchez, Y Zhang, E Le Chatelier, L Zitvogel, N Pons, J L Austin-Breneman, L E Haydu, E M Burton, J M Gardner, E Sirmans, J Hu, A J Lazar, T Tsujikawa, A Diab, H Tawbi, I C Glitza, W J Hwu, S P Patel, S E Woodman, R N Amaria, M A Davies, J E Gershenwald, P Hwu, J E Lee, J Zhang, L M Coussens, Z A Cooper, P A Futreal, C R Daniel, N J Ajami, J F Petrosino, M T Tetzlaff, P Sharma, J P Allison, R R Jenq, J A Wargo <<<

Preclinical mouse models suggest that the gut microbiome modulates tumor response to checkpoint blockade immunotherapy; however, this has not been well-characterized in human cancer patients. Here we examined the oral and gut microbiome of melanoma patients undergoing anti-programmed cell death 1 protein (PD-1) immunotherapy ( = 112). Significant differences were observed in the diversity and composition of the patient gut microbiome of responders versus nonresponders. Analysis of patient fecal microbiome samples ( = 43, 30 responders, 13 nonresponders) showed significantly higher alpha diversity ( < 0.01) and relative abundance of bacteria of the Ruminococcaceae family ( < 0.01) in responding patients. Metagenomic studies revealed functional differences in gut bacteria in responders, including enrichment of anabolic pathways. Immune profiling suggested enhanced systemic and antitumor immunity in responding patients with a favorable gut microbiome as well as in germ-free mice receiving fecal transplants from responding patients. Together, these data have important implications for the treatment of melanoma patients with immune checkpoint inhibitors. <<<

Junyue Cao, Malte Spielmann, Xiaojie Qiu, Xingfan Huang, Daniel M Ibrahim, Andrew J Hill, Fan Zhang, Stefan Mundlos, Lena Christiansen, Frank J Steemers, Cole Trapnell, Jay Shendure <<<

Mammalian organogenesis is a remarkable process. Within a short timeframe, the cells of the three germ layers transform into an embryo that includes most of the major internal and external organs. Here we investigate the transcriptional dynamics of mouse organogenesis at single-cell resolution. Using single-cell combinatorial indexing, we profiled the transcriptomes of around 2 million cells derived from 61 embryos staged between 9.5 and 13.5 days of gestation, in a single experiment. The resulting 'mouse organogenesis cell atlas' (MOCA) provides a global view of developmental processes during this critical window. We use Monocle 3 to identify hundreds of cell types and 56 trajectories, many of which are detected only because of the depth of cellular coverage, and collectively define thousands of corresponding marker genes. We explore the dynamics of gene expression within cell types and trajectories over time, including focused analyses of the apical ectodermal ridge, limb mesenchyme and skeletal muscle. <<<

Benoît Schmauch, Alberto Romagnoni, Elodie Pronier, Charlie Saillard, Pascale Maillé, Julien Calderaro, Aurélie Kamoun, Meriem Sefta, Sylvain Toldo, Mikhail Zaslavskiy, Thomas Clozel, Matahi Moarii, Pierre Courtiol, Gilles Wainrib <<<

Deep learning methods for digital pathology analysis are an effective way to address multiple clinical questions, from diagnosis to prediction of treatment outcomes. These methods have also been used to predict gene mutations from pathology images, but no comprehensive evaluation of their potential for extracting molecular features from histology slides has yet been performed. We show that HE2RNA, a model based on the integration of multiple data modes, can be trained to systematically predict RNA-Seq profiles from whole-slide images alone, without expert annotation. Through its interpretable design, HE2RNA provides virtual spatialization of gene expression, as validated by CD3- and CD20-staining on an independent dataset. The transcriptomic representation learned by HE2RNA can also be transferred on other datasets, even of small size, to increase prediction performance for specific molecular phenotypes. We illustrate the use of this approach in clinical diagnosis purposes such as the identification of tumors with microsatellite instability. <<<