RESUMO
Caregivers' touches that occur alongside words and utterances could aid in the detection of word/utterance boundaries and the mapping of word forms to word meanings. We examined changes in caregivers' use of touches with their speech directed to infants using a multimodal cross-sectional corpus of 35 Korean mother-child dyads across three age groups of infants (8, 14, and 27 months). We tested the hypothesis that caregivers' frequency and use of touches with speech change with infants' development. Results revealed that the frequency of word/utterance-touch alignment as well as word + touch co-occurrence is highest in speech addressed to the youngest group of infants. Thus, this study provides support for the hypothesis that caregivers' use of touch during dyadic interactions is sensitive to infants' age in a way similar to caregivers' use of speech alone and could provide cues useful to infants' language learning at critical points in early development.
Assuntos
Mães , Tato , Feminino , Humanos , Lactente , Estudos Transversais , Idioma , República da CoreiaRESUMO
Interest in computational modeling of cognition and behavior continues to grow. To be most productive, modelers should be equipped with tools that ensure optimal efficiency in data collection and in the integrity of inference about the phenomenon of interest. Traditionally, models in cognitive science have been parametric, which are particularly susceptible to model misspecification because their strong assumptions (e.g. parameterization, functional form) may introduce unjustified biases in data collection and inference. To address this issue, we propose a data-driven nonparametric framework for model development, one that also includes optimal experimental design as a goal. It combines Gaussian Processes, a stochastic process often used for regression and classification, with active learning, from machine learning, to iteratively fit the model and use it to optimize the design selection throughout the experiment. The approach, dubbed Gaussian process with active learning (GPAL), is an extension of the parametric, adaptive design optimization (ADO) framework (Cavagnaro, Myung, Pitt, & Kujala, 2010). We demonstrate the application and features of GPAL in a delay discounting task and compare its performance to ADO in two experiments. The results show that GPAL is a viable modeling framework that is noteworthy for its high sensitivity to individual differences, identifying novel patterns in the data that were missed by the model-constrained ADO. This investigation represents a first step towards the development of a data-driven cognitive modeling framework that serves as a middle ground between raw data, which can be difficult to interpret, and parametric models, which rely on strong assumptions.
Assuntos
Projetos de Pesquisa , Teorema de Bayes , Humanos , Distribuição Normal , Processos EstocásticosRESUMO
We describe a corpus of speech taking place between 30 Korean mother-child pairs, divided in three groups of Prelexical (M = 0;08), Early-Lexical (M = 1;02), and Advanced-Lexical (M = 2;03). In addition to the child-directed speech (CDS), this corpus includes two different formalities of adult-directed speech (ADS), i.e., family-directed ADS (ADS_Fam) and experimenter-directed ADS (ADS_Exp). Our analysis of the MLU in CDS, family-, and experimenter-directed ADS found significant differences between CDS and ADS_Fam, and between ADS_Fam and ADS_Exp, but not between CDS and ADS_Exp. Our finding suggests that researchers should pay more attention to controlling the level of formality in CDS and ADS when comparing the two registers for their speech characteristics. The corpus was transcribed in the CHAT format of the CHILDES system, so users can easily extract data related to verbal behavior in the mother-child interaction using the CLAN program of CHILDES.
RESUMO
Recent research in DNA nanotechnology has demonstrated that biological substrates can be used for computing at a molecular level. However, in vitro demonstrations of DNA computations use preprogrammed, rule-based methods which lack the adaptability that may be essential in developing molecular systems that function in dynamic environments. Here, we introduce an in vitro molecular algorithm that 'learns' molecular models from training data, opening the possibility of 'machine learning' in wet molecular systems. Our algorithm enables enzymatic weight update by targeting internal loop structures in DNA and ensemble learning, based on the hypernetwork model. This novel approach allows massively parallel processing of DNA with enzymes for specific structural selection for learning in an iterative manner. We also introduce an intuitive method of DNA data construction to dramatically reduce the number of unique DNA sequences needed to cover the large search space of feature sets. By combining molecular computing and machine learning the proposed algorithm makes a step closer to developing molecular computing technologies for future access to more intelligent molecular systems.
Assuntos
DNA , Aprendizado de Máquina , Modelos Moleculares , Redes Neurais de Computação , Conformação de Ácido Nucleico , DNA/química , DNA/genéticaRESUMO
BACKGROUND: Stress recognition using electrocardiogram (ECG) signals requires the intractable long-term heart rate variability (HRV) parameter extraction process. This study proposes a novel deep learning framework to recognize the stressful states, the Deep ECGNet, using ultra short-term raw ECG signals without any feature engineering methods. METHODS: The Deep ECGNet was developed through various experiments and analysis of ECG waveforms. We proposed the optimal recurrent and convolutional neural networks architecture, and also the optimal convolution filter length (related to the P, Q, R, S, and T wave durations of ECG) and pooling length (related to the heart beat period) based on the optimization experiments and analysis on the waveform characteristics of ECG signals. The experiments were also conducted with conventional methods using HRV parameters and frequency features as a benchmark test. The data used in this study were obtained from Kwangwoon University in Korea (13 subjects, Case 1) and KU Leuven University in Belgium (9 subjects, Case 2). Experiments were designed according to various experimental protocols to elicit stressful conditions. RESULTS: The proposed framework to recognize stress conditions, the Deep ECGNet, outperformed the conventional approaches with the highest accuracy of 87.39% for Case 1 and 73.96% for Case 2, respectively, that is, 16.22% and 10.98% improvements compared with those of the conventional HRV method. CONCLUSIONS: We proposed an optimal deep learning architecture and its parameters for stress recognition, and the theoretical consideration on how to design the deep learning structure based on the periodic patterns of the raw ECG data. Experimental results in this study have proved that the proposed deep learning model, the Deep ECGNet, is an optimal structure to recognize the stress conditions using ultra short-term ECG data.
Assuntos
Aprendizado Profundo , Eletrocardiografia/métodos , Processamento de Imagem Assistida por Computador/métodos , Estresse Psicológico/fisiopatologia , Adulto , Bélgica , Frequência Cardíaca/fisiologia , Humanos , Masculino , Redes Neurais de Computação , República da Coreia , Adulto JovemRESUMO
Classical discriminant analysis attempts to discover a low-dimensional subspace where class label information is maximally preserved under projection. Canonical methods for estimating the subspace optimize an information-theoretic criterion that measures the separation between the class-conditional distributions. Unfortunately, direct optimization of the information-theoretic criteria is generally non-convex and intractable in high-dimensional spaces. In this work, we propose a novel, tractable algorithm for discriminant analysis that considers the class-conditional densities as interacting fluids in the high-dimensional embedding space. We use the Bhattacharyya criterion as a potential function that generates forces between the interacting fluids, and derive a computationally tractable method for finding the low-dimensional subspace that optimally constrains the resulting fluid flow. We show that this model properly reduces to the optimal solution for homoscedastic data as well as for heteroscedastic Gaussian distributions with equal means. We also extend this model to discover optimal filters for discriminating Gaussian processes and provide experimental results and comparisons on a number of datasets.
RESUMO
We consider the problem of learning a local metric in order to enhance the performance of nearest neighbor classification. Conventional metric learning methods attempt to separate data distributions in a purely discriminative manner; here we show how to take advantage of information from parametric generative models. We focus on the bias in the information-theoretic error arising from finite sampling effects, and find an appropriate local metric that maximally reduces the bias based upon knowledge from generative models. As a byproduct, the asymptotic theoretical analysis in this work relates metric learning to dimensionality reduction from a novel perspective, which was not understood from previous discriminative approaches. Empirical experiments show that this learned local metric enhances the discriminative nearest neighbor performance on various datasets using simple class conditional generative models such as a Gaussian.
RESUMO
Programmable biomolecules, such as DNA strands, deoxyribozymes, and restriction enzymes, have been used to solve computational problems, construct large-scale logic circuits, and program simple molecular games. Although studies have shown the potential of molecular computing, the capability of computational learning with DNA molecules, i.e., molecular machine learning, has yet to be experimentally verified. Here, we present a novel molecular learning in vitro model in which symmetric internal loops of double-stranded DNA are exploited to measure the differences between training instances, thus enabling the molecules to learn from small errors. The model was evaluated on a data set of twenty dialogue sentences obtained from the television shows Friends and Prison Break. The wet DNA-computing experiments confirmed that the molecular learning machine was able to generalize the dialogue patterns of each show and successfully identify the show from which the sentences originated. The molecular machine learning model described here opens the way for solving machine learning problems in computer science and biology using in vitro molecular computing with the data encoded in DNA molecules.
Assuntos
Algoritmos , DNA , Simulação de Dinâmica Molecular , Animais , Humanos , Lógica , Aprendizado de MáquinaRESUMO
Wearable devices, such as smart glasses and watches, allow for continuous recording of everyday life in a real world over an extended period of time or lifelong. This possibility helps better understand the cognitive behavior of humans in real life as well as build human-aware intelligent agents for practical purposes. However, modeling the human cognitive activity from wearable-sensor data stream is challenging because learning new information often results in loss of previously acquired information, causing a problem known as catastrophic forgetting. Here we propose a deep-learning neural network architecture that resolves the catastrophic forgetting problem. Based on the neurocognitive theory of the complementary learning systems of the neocortex and hippocampus, we introduce a dual memory architecture (DMA) that, on one hand, slowly acquires the structured knowledge representations and, on the other hand, rapidly learns the specifics of individual experiences. The DMA system learns continuously through incremental feature adaptation and weight transfer. We evaluate the performance on two real-life datasets, the CIFAR-10 image-stream dataset and the 46-day Lifelog dataset collected from Google Glass, showing that the proposed model outperforms other online learning methods.
Assuntos
Cognição , Microcomputadores , Modelos Neurológicos , Redes Neurais de Computação , Encéfalo/fisiologia , HumanosRESUMO
We present a computational learning method for bio-molecular classification. This method shows how to design biochemical operations both for learning and pattern classification. As opposed to prior work, our molecular algorithm learns generic classes considering the realization in vitro via a sequence of molecular biological operations on sets of DNA examples. Specifically, hybridization between DNA molecules is interpreted as computing the inner product between embedded vectors in a corresponding vector space, and our algorithm performs learning of a binary classifier in this vector space. We analyze the thermodynamic behavior of these learning algorithms, and show simulations on artificial and real datasets as well as demonstrate preliminary wet experimental results using gel electrophoresis.
Assuntos
DNA/genética , Vetores Genéticos , Hibridização de Ácido Nucleico , TermodinâmicaRESUMO
Although the regulation of membrane receptor activation is known to be crucial for molecular signal transduction, the molecular mechanism underlying receptor activation is not fully elucidated. Here we study the physicochemical nature of membrane receptor behavior by investigating the characteristic molecular vibrations of receptor ligands using computational chemistry and informatics methods. By using information gain, t-tests, and support vector machines, we have identified highly informative features of adenosine receptor (AdoR) ligand and corresponding functional amino acid residues such as Asn (6.55) of AdoR that has informative significance and is indispensable for ligand recognition of AdoRs. These findings may provide new perspectives and insights into the fundamental mechanism of class A G protein-coupled receptor activation.
Assuntos
Agonistas do Receptor Purinérgico P1/química , Antagonistas de Receptores Purinérgicos P1/química , Humanos , Ligantes , Modelos Moleculares , Ligação Proteica , Teoria Quântica , Receptores Purinérgicos P1/químicaRESUMO
Predicting the clinical outcomes of cancer patients is a challenging task in biomedicine. A personalized and refined therapy based on predicting prognostic outcomes of cancer patients has been actively sought in the past decade. Accurate prognostic prediction requires higher-order representations of complex dependencies among genetic factors. However, identifying the co-regulatory roles and functional effects of genetic interactions on cancer prognosis is hindered by the complexity of the interactions. Here we propose a prognostic prediction model based on evolutionary learning that identifies higher-order prognostic biomarkers of cancer clinical outcomes. The proposed model represents the interactions of prognostic genes as a combinatorial space. It adopts a flexible hypergraph structure composed of a large population of hyperedges that encode higher-order relationships among many genetic factors. The hyperedge population is optimized by an evolutionary learning method based on sequential Bayesian sampling. The proposed learning approach effectively balances performance and parsimony of the model using information-theoretic dependency and complexity-theoretic regularization priors. Using MAQC-II project data, we demonstrate that our model can handle high-dimensional data more effectively than state-of-the-art classification models. We also identify potential gene interactions characterizing prognosis and recurrence risk in cancer.
Assuntos
Teorema de Bayes , Aprendizagem , Neoplasias/terapia , Humanos , Neoplasias/patologia , Resultado do TratamentoRESUMO
MicroRNAs (miRNAs) are small regulatory molecules that repress the translational processes of their target genes by binding to their 3' untranslated regions (3' UTRs). Because the target genes are predominantly determined by their sequence complementarity to the miRNA seed regions (nucleotides 2-7) which are evolutionarily conserved, it is inferred that the target relationships and functions of the miRNA family members are conserved across many species. Therefore, detecting the relevant miRNA families with confidence would help to clarify the conserved miRNA functions, and elucidate miRNA-mediated biological processes. We present a mixture model of position weight matrices for constructing miRNA functional families. This model systematically finds not only evolutionarily conserved miRNA family members but also functionally related miRNAs, as it simultaneously generates position weight matrices representing the conserved sequences. Using mammalian miRNA sequences, in our experiments, we identified potential miRNA groups characterized by similar sequence patterns that have common functions. We validated our results using score measures and by the analysis of the conserved targets. Our method would provide a way to comprehensively identify conserved miRNA functions.
RESUMO
Aberrant DNA methylation of CpG islands, CpG island shores and first exons is known to play a key role in the altered gene expression patterns in all human cancers. To date, a systematic study on the effect of DNA methylation on gene expression using high resolution data has not been reported. In this study, we conducted an integrated analysis of MethylCap-sequencing data and Affymetrix gene expression microarray data for 30 breast cancer cell lines representing different breast tumor phenotypes. As well-developed methods for the integrated analysis do not currently exist, we created a series of four different analysis methods. On the computational side, our goal is to develop methylome data analysis protocols for the integrated analysis of DNA methylation and gene expression data on the genome scale. On the cancer biology side, we present comprehensive genome-wide methylome analysis results for differentially methylated regions and their potential effect on gene expression in 30 breast cancer cell lines representing three molecular phenotypes, luminal, basal A and basal B. Our integrated analysis demonstrates that methylation status of different genomic regions may play a key role in establishing transcriptional patterns in molecular subtypes of human breast cancer.
Assuntos
Neoplasias da Mama/genética , Metilação de DNA , Regulação Neoplásica da Expressão Gênica , Sítios de Ligação , Neoplasias da Mama/classificação , Neoplasias da Mama/metabolismo , Linhagem Celular Tumoral , Regulação para Baixo , Feminino , Perfilação da Expressão Gênica , Genômica/métodos , Humanos , Fenótipo , Regiões Promotoras Genéticas , Fatores de Transcrição/metabolismoRESUMO
In vitro pattern classification has been highlighted as an important future application of DNA computing. Previous work has demonstrated the feasibility of linear classifiers using DNA-based molecular computing. However, complex tasks require non-linear classification capability. Here we design a molecular beacon that can interact with multiple targets and experimentally shows that its fluorescent signals form a complex radial-basis function, enabling it to be used as a building block for non-linear molecular classification in vitro. The proposed method was successfully applied to solving artificial and real-world classification problems: XOR and microRNA expression patterns.
Assuntos
Classificação/métodos , Computadores Moleculares/tendências , MicroRNAs/metabolismo , Modelos Teóricos , Fluorescência , MicroRNAs/classificação , MicroRNAs/genéticaRESUMO
BACKGROUND: Dysregulation of genetic factors such as microRNAs (miRNAs) and mRNAs has been widely shown to be associated with cancer progression and development. In particular, miRNAs and mRNAs cooperate to affect biological processes, including tumorigenesis. The complexity of miRNA-mRNA interactions presents a major barrier to identifying their co-regulatory roles and functional effects. Thus, by computationally modeling these complex relationships, it may be possible to infer the gene interaction networks underlying complicated biological processes. RESULTS: We propose a data-driven, hypergraph structural method for constructing higher-order miRNA-mRNA interaction networks from cancer genomic profiles. The proposed model explicitly characterizes higher-order relationships among genetic factors, from which cooperative gene activities in biological processes may be identified. The proposed model is learned by iteration of structure and parameter learning. The structure learning efficiently constructs a hypergraph structure by generating putative hyperedges representing complex miRNA-mRNA modules. It adopts an evolutionary method based on information-theoretic criteria. In the parameter learning phase, the constructed hypergraph is refined by updating the hyperedge weights using the gradient descent method. From the model, we produce biologically relevant higher-order interaction networks showing the properties of primary and metastatic prostate cancer, as candidates of potential miRNA-mRNA regulatory circuits. CONCLUSIONS: Our approach focuses on potential cancer-specific interactions reflecting higher-order relationships between miRNAs and mRNAs from expression profiles. The constructed miRNA-mRNA interaction networks show oncogenic or tumor suppression characteristics, which are known to be directly associated with prostate cancer progression. Therefore, the hypergraph-based model can assist hypothesis formulation for the molecular pathogenesis of cancer.
Assuntos
Inteligência Artificial , Biologia Computacional/métodos , Gráficos por Computador , Redes Reguladoras de Genes , MicroRNAs/genética , Neoplasias da Próstata/genética , Humanos , Masculino , RNA Mensageiro/genéticaRESUMO
Molecular beacons are efficient and useful tools for quantitative detection of specific target nucleic acids. Thanks to their simple protocol, molecular beacons have great potential as substrates for biomolecular computing. Here we present a molecular beacon-based biomolecular computing method for quantitative detection and analysis of target nucleic acids. Whereas the conventional quantitative assays using fluorescent dyes have been designed for single target detection or multiplexed detection, the proposed method enables us not only to detect multiple targets but also to compute their quantitative information by weighted-sum of the targets. The detection and computation are performed on a molecular level simultaneously, and the outputs are detected as fluorescence signals. Experimental results show the feasibility and effectiveness of our weighted detection and linear combination method using molecular beacons. Our method can serve as a primitive operation of molecular pattern analysis, and we demonstrate successful binary classifications of molecular patterns made of synthetic oligonucleotide DNA molecules.
Assuntos
Biologia Computacional/métodos , Sondas de DNA/metabolismo , Ácidos Nucleicos/análise , Sequência de Bases , Sondas de DNA/genética , Fluorescência , Dados de Sequência Molecular , Oligonucleotídeos/genéticaRESUMO
Biomolecules inside a microfluidic system can be used to solve computational problems, such as theorem proving, which is an important class of logical reasoning problems. In this article, the Boolean variables (literals) were represented using single-stranded DNA molecules, and theorem proving was performed by the hybridization and ligation of these variables into a double-stranded "solution" DNA. Then, a novel sequential reaction mixing method in a microfluidic chip was designed to solve a theorem proving problem, where a reaction loop and three additional chambers were integrated and controlled by pneumatic valves. DNA hybridization, ligation, toehold-mediated DNA strand displacement, exonuclease I digestion, and fluorescence detection of the double-stranded DNA were sequentially performed using this platform. Depending on the computational result, detection of the correct answer was demonstrated based on the presence of a fluorescence signal. This result is the first demonstration that microfluidics can be used to facilitate DNA-based logical inference.
Assuntos
Computadores Moleculares , DNA de Cadeia Simples/química , Técnicas Analíticas Microfluídicas/instrumentação , Técnicas Analíticas Microfluídicas/métodos , Modelos Químicos , Carbocianinas/química , Desenho de Equipamento , Corantes Fluorescentes/química , Lógica , Espectrometria de Fluorescência/métodosRESUMO
BACKGROUND: Biclustering has been utilized to find functionally important patterns in biological problem. Here a bicluster is a submatrix that consists of a subset of rows and a subset of columns in a matrix, and contains homogeneous patterns. The problem of finding biclusters is still challengeable due to computational complex trying to capture patterns from two-dimensional features. RESULTS: We propose a Probabilistic COevolutionary Biclustering Algorithm (PCOBA) that can cluster the rows and columns in a matrix simultaneously by utilizing a dynamic adaptation of multiple species and adopting probabilistic learning. In biclustering problems, a coevolutionary search is suitable since it can optimize interdependent subcomponents formed of rows and columns. Furthermore, acquiring statistical information on two populations using probabilistic learning can improve the ability of search towards the optimum value. We evaluated the performance of PCOBA on synthetic dataset and yeast expression profiles. The results demonstrated that PCOBA outperformed previous evolutionary computation methods as well as other biclustering methods. CONCLUSIONS: Our approach for searching particular biological patterns could be valuable for systematically understanding functional relationships between genes and other biological components at a genome-wide level.
Assuntos
Perfilação da Expressão Gênica/estatística & dados numéricos , Análise de Sequência com Séries de Oligonucleotídeos/estatística & dados numéricos , Algoritmos , Análise por Conglomerados , Evolução Molecular , Expressão Gênica , Saccharomyces cerevisiae/genéticaRESUMO
Recent results of corpus-based linguistics demonstrate that context-appropriate sentences can be generated by a stochastic constraint satisfaction process. Exploiting the similarity of constraint satisfaction and DNA self-assembly, we explore a DNA assembly model of sentence generation. The words and phrases in a language corpus are encoded as DNA molecules to build a language model of the corpus. Given a seed word, the new sentences are constructed by a parallel DNA assembly process based on the probability distribution of the word and phrase molecules. Here, we present our DNA code word design and report on successful demonstration of their feasibility in wet DNA experiments of a small scale.