Inference Relations Research Articles

Inferring ongoing cancer evolution from single tumour biopsies using synthetic supervised learning.

Variant allele frequencies (VAF) encode ongoing evolution and subclonal selection in growing tumours. However, existing methods that utilize VAF information for cancer evolutionary inference are compressive, slow, or incorrectly specify the underlying cancer evolutionary dynamics. Here, we provide a proof-of-principle synthetic supervised learning method, TumE, that integrates simulated models of cancer evolution with Bayesian neural networks, to infer ongoing selection in bulk-sequenced single tumour biopsies. Analyses in synthetic and patient tumours show that TumE significantly improves both accuracy and inference time per sample when detecting positive selection, deconvoluting selected subclonal populations, and estimating subclone frequency. Importantly, we show how transfer learning can leverage stored knowledge within TumE models for related evolutionary inference tasks-substantially reducing data and computational time for further model development and providing a library of recyclable deep learning models for the cancer evolution community. This extensible framework provides a foundation and future directions for harnessing progressive computational methods for the benefit of cancer genomics and, in turn, the cancer patient.

Inferences Between Buridan’s Modal Propositions

In recent years modal syllogistic provided by 14th century logician John Buridan has attracted increasing attention of historians of medieval logic. The widespread use of quantified modal logic with the apparatus of possible worlds semantics in current analytic philosophy has encouraged the investigation of the relation of Buridan’s theory of modality with the modern developments of symbolic modal logic. We focus on the semantics of and the inferential relations among the propositions that underlie Buridan’s theory of modal syllogism. First, we review all inferences between propositions of necessity, possibility, contingency, and non-contingency, with or without quod est locution, that are valid in Buridan’s semantics, and offer a comprehensive diagrammatic representation that includes them all. We then ask the question if there is a way to model those results in first order modal logic. Three ways of formalizing Buridan’s propositions in quantified modal logic are considered. Comparison of inferences between the quantified formulas and Buridan’s propositions reveals that, when supplied with a suitable formalization, Buridan’s semantics of categorical statements and immediate inferences among them can be fully captured by the quantified modal system T.

Computational causal discovery: Advantages and assumptions

I would like to congratulate James Woodward for another landmark accomplishment, after publishing his Making things happen: A theory of causal explanation (Woodward, 2003). Making things happen gives an elegant interventionist theory for understanding explanation and causation. The new contribution (Woodward, 2022) relies on that theory and further makes a big step towards empirical inference of causal relations from non-experimental data. In this paper, I will focus on some of the emerging computational methods for finding causal relations from non-experimental data and attempt to complement Woodward's contribution with discussions on 1) how these methods are connected to the interventionist theory of causality, 2) how informative the output of the methods is, including whether they output directed causal graphs and how they deal with confounders (unmeasured common causes of two measured variables), and 3) the assumptions underlying the asymptotic correctness of the output of the methods about causal relations. Different causal discovery methods may rely on different aspects of the joint distribution of the data, and this discussion aims to provide a technical account of the assumptions.

Neural relational inference to learn long-range allosteric interactions in proteins from molecular dynamics simulations

Protein allostery is a biological process facilitated by spatially long-range intra-protein communication, whereby ligand binding or amino acid change at a distant site affects the active site remotely. Molecular dynamics (MD) simulation provides a powerful computational approach to probe the allosteric effect. However, current MD simulations cannot reach the time scales of whole allosteric processes. The advent of deep learning made it possible to evaluate both spatially short and long-range communications for understanding allostery. For this purpose, we applied a neural relational inference model based on a graph neural network, which adopts an encoder-decoder architecture to simultaneously infer latent interactions for probing protein allosteric processes as dynamic networks of interacting residues. From the MD trajectories, this model successfully learned the long-range interactions and pathways that can mediate the allosteric communications between distant sites in the Pin1, SOD1, and MEK1 systems. Furthermore, the model can discover allostery-related interactions earlier in the MD simulation trajectories and predict relative free energy changes upon mutations more accurately than other methods.

In vitro Inhibition of respiratory pathogens by lactobacillus and alpha haemolytic streptococci from Aboriginal and Torres Strait Islander children.

To explore the in vitro ability of alpha haemolytic streptococcus (AHS) and lactobacilli (LBs), from Indigenous Australian children, to inhibit the growth of respiratory pathogens (Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis), also from Indigenous Australian children. The bacterial interference of 91 isolates, from Indigenous Australian children both with and without otitis media (OM) or rhinorrhoea, was investigated using agar overlay and cell-free supernatant. Promising isolates underwent whole genome sequencing to investigate upper respiratory tract tropism, antibiotic resistance and virulence. Antibiotic susceptibility was examined for ampicillin, amoxicillin +clavulanic acid and azithromycin. Differences in the strength of bacterial inferences in relation to OM was examined using a case series of three healthy and three children with OM. LBs readily inhibited the growth of pathogens. AHS were less effective, although several isolates inhibited S. pneumoniae. One L. rhamnosus had genes coding for pili to adhere to epithelial cells. We detected antibiotic resistance genes coding for antibiotic efflux pump and ribosomal protection protein. LBs were susceptible to antimicrobials in vitro. Screening for virulence detected genes encoding for two putative capsule proteins. Healthy children had AHS and LB that were more potent inhibitors of respiratory pathogens in vitro than children with OM. L. rhamnosus from remote Indigenous Australian children are potent inhibitors of respiratory pathogens in vitro. Respiratory/ear disease are endemic in Indigenous Australians. There is an urgent call for more effective treatment/prevention; beneficial microbes have not been explored. L. rhamnosus investigated in this study are potent inhibitors of respiratory pathogens in vitro and require further investigation.

A Shewhart-type nonparametric multivariate depth-based control chart for monitoring location

Parametric multivariate control charts like Hotelling’s rely on certain distributional assumptions that if these assumptions are not met, it leads us to an excessive number of false alarms and reduces the effectiveness of the monitoring strategy. In contrast, nonparametric multivariate control charts, specifically data-depth-based charts and related inferences, do not require any distributional assumptions and no constraints on data dimensions. In this paper, based on a family of multivariate affine invariant distribution-free data-depth-based tests for the multivariate one-sample location problem, a Shewhart-type nonparametric multivariate phase-II control chart is proposed. The performance of the proposed control chart is evaluated with a Monte Carlo simulation study using the average run length measure and is compared to that of competitors. It is observed that the proposed control chart maintained a desirable performance and leads to quicker detection of shifts. It is also a close competitor to Hotelling’s chart when the underlying distribution of the process is bivariate normal. Moreover, it performs better than its nonparametric competitor under the bivariate non-normal elliptical direction class of distributions, specifically heavy-tailed, light-tailed, and t-distributions, in a variety of settings. The control chart procedure is illustrated as an application to real-life data.

Asymmetric reinforcement learning facilitates human inference of transitive relations

Humans and other animals are capable of inferring never-experienced relations (for example, A > C) from other relational observations (for example, A > B and B > C). The processes behind such transitive inference are subject to intense research. Here we demonstrate a new aspect of relational learning, building on previous evidence that transitive inference can be accomplished through simple reinforcement learning mechanisms. We show in simulations that inference of novel relations benefits from an asymmetric learning policy, where observers update only their belief about the winner (or loser) in a pair. Across four experiments (n = 145), we find substantial empirical support for such asymmetries in inferential learning. The learning policy favoured by our simulations and experiments gives rise to a compression of values that is routinely observed in psychophysics and behavioural economics. In other words, a seemingly biased learning strategy that yields well-known cognitive distortions can be beneficial for transitive inferential judgements.

Techniques for reconstruction after distal gastrectomy for cancer: updated network meta-analysis of randomized controlled trials.

The choice of the best reconstruction technique after distal gastrectomy (DG) remains controversial and still not defined. The purpose was to perform a comprehensive evaluation within the major type of intestinal reconstruction after DG for gastric cancer. Systematic review and network meta-analyses of randomized controlled trials (RCTs) to compare Billroth I (BI), Billroth II (BII), Billroth II Braun (BII Braun), Roux-en-Y (RY), and Uncut Roux-en-Y (URY). Risk ratio (RR) and weighted mean difference (WMD) were used as pooled effect size measures while 95% credible intervals (CrI) were used to assess relative inference. Ten RCTs (1456 patients) were included. Of these, 448 (33.7%) underwent BI, 220 (15.1%) BII, 114 BII Braun (7.8%), 533 (36.6%) RY, and 141 URY (9.6%). No significant differences were found among treatments for 30-day mortality, anastomotic leak, anastomotic stricture, and overall complications. At 12-month follow-up, RY was associated with a significantly reduced risk of remnant gastritis compared to BI (RR=0.56; 95% Crl 0.35-0.76) and BII reconstruction (RR=0.47; 95% Crl 0.22-0.97). Similarly, despite the lack of statistical significance, RY seems associated with a trend toward reduced endoscopically proven esophagitis compared to BI (RR=0.58; 95% Crl 0.24-1.51) and bile reflux compared to BI (RR=0.48; 95% Crl 0.17-1.41), BII (RR=0.74; 95% Crl 0.20-2.81), and BII Braun (RR=0.65; 95% Crl 0.30-1.43). This network meta-analysis shows that there are five main options for intestinal anastomosis after DG. All techniques seem equally safe with comparable anastomotic leak, anastomotic stricture, overall morbidity, and short-term outcomes. In the short-term follow-up (12 months), RY seems associated with a reduced risk of remnant gastritis and a trend toward a reduced risk of bile reflux and esophagitis.

ProSGPV: an R package for variable selection with second-generation p-values

We introduce the ProSGPV R package, which implements a variable selection algorithm based on second-generation p-values (SGPV) instead of traditional p-values. Most variable selection algorithms shrink point estimates to arrive at a sparse solution. In contrast, the ProSGPV algorithm accounts for the estimation uncertainty – via confidence intervals – in the selection process. This additional information leads to better inference and prediction performance in finite sample sizes. ProSGPV maintains good performance even in the high dimensional case where $p>n$, or when explanatory variables are highly correlated. Moreover, ProSGPV is a unifying algorithm that works with continuous, binary, count, and time-to-event outcomes. No cross-validation or iterative processes are needed and thus ProSGPV is very fast to compute. Visualization tools are available in this package for assessing the variable selection process. Here we present simulation studies and a real-world example to demonstrate ProSGPV’s inference and prediction performance in relation to the current standards in variable selection procedures.

Relationship Between Replay-Associated Ripples and Hippocampal N-Methyl-D-Aspartate Receptors: Preliminary Evidence From a PET-MEG Study in Schizophrenia.

Background and HypothesesHippocampal replay and associated high-frequency ripple oscillations are among the best-characterized phenomena in resting brain activity. Replay/ripples support memory consolidation and relational inference, and are regulated by N-methyl-D-aspartate receptors (NMDARs). Schizophrenia has been associated with both replay/ripple abnormalities and NMDAR hypofunction in both clinical samples and genetic mouse models, although the relationship between these 2 facets of hippocampal function has not been tested in humans. Study DesignHere, we avail of a unique multimodal human neuroimaging data set to investigate the relationship between the availability of (intrachannel) NMDAR binding sites in hippocampus, and replay-associated ripple power, in 16 participants (7 nonclinical participants and 9 people with a diagnosis of schizophrenia, PScz). Each participant had both a [18F]GE-179 positron emission tomography (PET) scan (to measure NMDAR availability, VT) and a magnetoencephalography (MEG) scan (to measure offline neural replay and associated high-frequency ripple oscillations, using Temporally Delayed Linear Modeling). Study ResultsWe show a positive relationship between hippocampal NMDAR availability and replay-associated ripple power. This linkage was evident across control participants (r(5) = .94, P = .002) and PScz (r(7) = .70, P = .04), with no group difference. ConclusionsOur findings provide preliminary evidence for a relationship between hippocampal NMDAR availability and replay-associated ripple power in humans, and haverelevance for NMDAR hypofunction theories of schizophrenia.

Transformer-Driven Semantic Relation Inference for Multilabel Classification of High-Resolution Remote Sensing Images

It is hard to use a single label to describe an image for the complexity of remote sensing scenes. Thus, it is a more general and practical choice to use multilabel image classification for high-resolution remote sensing (HRS) images. How to construct the relation between categories is a vital problem for multilabel classification. Some researchers use the recurrent neural network (RNN) or long short-term memory (LSTM) to exploit label relations over the last years. However, the RNN or LSTM could model such category dependence in a chain propagation manner. The performance of the RNN/LSTM might be questioned when a specific category is improperly inferred. To address this, we propose a novel HRS image multilabel classification network, transformer-driven semantic relation inference network. The network comprises two modules: semantic sensitive module (SSM) and semantic relation-building module (SRBM). The SSM locates the semantic attentional regions in the features extracted by a deep convolutional neural network and generates a discriminative content-aware category representation (CACR). The SRBM uses label relation inference from outputs of the SSM to predict final results. The characteristic of the proposed method is that it can extract semantic attentional regions relevant to the category and generate a discriminative CACR and natural and interpretable reasoning about label relations. Experiments were performed on the public UCM multilabel and MLRSNet datasets. Quantitative and qualitative analyses on state-of-the-art multilabel benchmarks proved that the proposed method could effectively locate semantic regions and build relationships between categories with better robustness.

A method for evaluating the relationship strength of group users based on co-occurrence

At present, the inference of users' social relations based on check-in data has become a hot topic. However, the existing methods to evaluate users' relationship strength mainly concentrated between two users, and group users contain greater information and utilisation value. This paper focuses on the co-occurrence phenomenon that group users are in the same time and space, and research group users' relationship strength problem with co-occurrence method and min-max method based on entropy. The real check-in data sets are used to carry out sufficient experiments on the two proposed evaluation methods. The experimental results show that the min-max method can better adapt to the sparse problem of check-in data and reduce the influence of coincidence. Compared with the basic co-occurrence method, it has certain stability in groups of different sizes, and the larger the group size is, the more obvious its advantages are.

Latent Space Semantic Supervision Based on Knowledge Distillation for Cross-Modal Retrieval.

As an important field in information retrieval, fine-grained cross-modal retrieval has received great attentions from researchers. Existing fine-grained cross-modal retrieval methods made several improvements in capturing the fine-grained interplay between vision and language, failing to consider the fine-grained correspondences between the features in the image latent space and the text latent space respectively, which may lead to inaccurate inference of intra-modal relations or false alignment of cross-modal information. Considering that object detection can get the fine-grained correspondences of image region features and the corresponding semantic features, this paper proposed a novel latent space semantic supervision model based on knowledge distillation (L3S-KD), which trains classifiers supervised by the fine-grained correspondences obtained from an object detection model by using knowledge distillation for image latent space fine-grained alignment, and by the labels of objects and attributes for text latent space fine-grained alignment. Compared with existing fine-grained correspondence matching methods, L3S-KD can learn more accurate semantic similarities for local fragments in image-text pairs. Extensive experiments on MS-COCO and Flickr30K datasets demonstrate that the L3S-KD model consistently outperforms state-of-the-art methods for image-text matching.

Nyāya Formalized: Exercises of Application

This article aims to provide some basic examples of the application of 'Navya-Nyāya Formal Language' (NL). NL consists in a method for formalizing Navya-Nyāya technical language and inferences. Derived especially from the works of Matilal, Staal, and Ganeri and conceived primarily as a hermeneutical tool, NL expands the expressive power of its forerunners, in particular in terms of handling complex relations and remaining faithful to the original Navya-Nyāya–style Sanskrit formulations. An extensional set-based approach and non-predicative relational syntax are adopted here, proposing solutions to deal with quantification shift. A thumbnail sketch of a relational formalistic inference is also provided.

Nature and Norm: Judaism, Christianity, and the Theopolitical Problem

Nature and Norm: Judaism, Christianity, and the Theopolitical Problem

Research on Event Logic Knowledge Graph Construction Method of Robot Transmission System Fault Diagnosis

Knowledge graph technology has important guiding significance for efficient and orderly fault diagnosis of robot transmission system. Taking the historical robot maintenance logs of robot transmission system as the research object, a top-down fault diagnosis event logic knowledge graph construction method is proposed. Firstly, we define event arguments of fault phenomenon and fault cause events, define event argument classes and relation between classes, and construct an event logic knowledge ontology model. According to the event logic knowledge ontology, the fault diagnosis event argument entity and relation in the corpus are labeled, and an event logic knowledge extraction dataset is formed. Secondly, an event argument entity and relation joint extraction model is proposed. Using stacked bidirectional long short-term memory(BiLSTM) to obtain deep context features of text. As a supplement to stacked BiLSTM, self-attention mechanism extracts character dependency features from multiple subspaces, and uses conditional random field(CRF) to realize entity recognition. The character dependency features are mapped to the entity label weight embedding, and spliced with deep context features to extract relations. Bidirectional graph convolutional network(BiGCN) is introduced for relation inference, graph convolution features are used to update deep context features to perform joint extraction in the second phase. Experimental results show that this method can improve the effect of event argument entity and relation joint extraction and is better than other methods. Finally, an event logic knowledge graph of robot transmission system fault diagnosis is constructed, which provides decision support for autonomous fault diagnosis of robot transmission system.

Variational Deep Logic Network for Joint Inference of Entities and Relations

AbstractCurrently, deep learning models have been widely adopted and achieved promising results on various application domains. Despite their intriguing performance, most deep learning models function as black boxes, lacking explicit reasoning capabilities and explanations, which are usually essential for complex problems. Take joint inference in information extraction as an example. This task requires the identification of multiple structured knowledge from texts, which is inter-correlated, including entities, events, and the relationships between them. Various deep neural networks have been proposed to jointly perform entity extraction and relation prediction, which only propagate information implicitly via representation learning. However, they fail to encode the intensive correlations between entity types and relations to enforce their coexistence. On the other hand, some approaches adopt rules to explicitly constrain certain relational facts, although the separation of rules with representation learning usually restrains the approaches with error propagation. Moreover, the predefined rules are inflexible and might result in negative effects when data is noisy. To address these limitations, we propose a variational deep logic network that incorporates both representation learning and relational reasoning via the variational EM algorithm. The model consists of a deep neural network to learn high-level features with implicit interactions via the self-attention mechanism and a relational logic network to explicitly exploit target interactions. These two components are trained interactively to bring the best of both worlds. We conduct extensive experiments ranging from fine-grained sentiment terms extraction, end-to-end relation prediction, to end-to-end event extraction to demonstrate the effectiveness of our proposed method.

Studying insecurity from relative safety — Dealing with methodological blind spots

Qualitative empirical enquiries into dynamics of security and insecurity often include a blind spot that bear theoretical ramifications because only those areas and respondents that allow for relatively safe fieldwork are studied. To transparently articulate the spheres of projection that creep into our knowledge production, we propose a distinction between inner and outer circles as highly fluid but separate geographical, socio-political and methodological spaces. Drawing on fieldwork in the Central African Republic and South Sudan, we discuss the risks posed by incomplete data and subsequently flawed inferences. We argue that the perceptions of fear projected onto the outer circle shape people’s behaviour more than measurable insecurity incidents and that increased interaction between actors in both circles reduces the perceived threats coming from the outer circles. We demonstrate how studying insecurity from inner circles risks securitizing outer circles while further centralizing the inner ones. We thus urge transparency in data collection and the related inferences that underpin our knowledge production.

Relative Inference for Paired Data: More than Meets the Eye

Relative Inference for Paired Data: More than Meets the Eye

Robust Covariance Matrix Estimation in Time Series: A Review

In the analysis of economic, financial and other time series, long-run variance estimators play an important role in estimating model parameters more efficiently and drawing more accurate statistical inference on the parameters. A non-technical review of long-run variance estimation is provided. Both parametric and nonparametric estimators are discussed. Kernel methods are dominant among all estimation procedures, and therefore recent developments in kernel-smoothed estimators and related inference are presented. The information given can help practitioners decide on a suitable long-run variance estimator.