基于深度玻尔兹曼机的蛋白质相互作用预测

本文针对传统蛋白质相互作用预测模型预测精度不够高的问题,提出一种改进的深度玻尔兹曼机(DBM)模型以更精确地预测蛋白质的相互作用。首先,将多尺度特征组提取和自协方差编码方法结合编码序列特征,并利用DBM自动筛选有效特征。同时,为了避免采用sigmoid或tanh激活函数在深度网络中出现过饱和的问题,本文采用Re LU改进的深度玻尔兹曼机(RBM),使网络具备稀疏性,从而避免模型过拟合,加快收敛速度。在酵母菌PPIs数据集上,本文算法达到了92.27%的准确率,优于传统的方法。     

具有节点偏置的高阶神经网络模型

在汪涛文献基础上提出了一个具有节点偏置的高阶神经网络模型、给出了模型的哈密顿量和学习算法,证明了学习算法的收敛性,该模型能对每一神经元自动引入一个节点偏置使得网络能够存储所有学习图样包括相关图样,其存储容量远高于Hebb—rule—like学习算法下的高阶神经网络模型.对由30个神经元组成的二阶神经网络进行了计算机仿真,结果证实了上述结论.此外,对初始突触强度对学习效果的影响和不同存储图样数目下的平均吸引半径进行了仿真计算并分析了所得结果.新模型的特点使其具有良好的应用前景     

扩散模式对高阶相互作用物种共存机制的影响

物种共存机制是群落生态学研究的核心问题之一,但以成对物种间直接相互作用为主的传统共存理论,并未在实际群落中得到普遍证实。近年来,有研究表明,高阶相互作用,即一个物种对另一个物种的直接作用强度受到其他物种的间接影响,在群落竞争过程中的重要性不断得到发展。目前,对高阶相互作用的理论研究还主要集中在非空间理论模型。事实上,群落中个体的空间分布和扩散模式等对种群动态的影响均至关重要。故考虑空间因素,以三物种为例构建空间显式的群落动态模拟,通过引入不同的物种扩散模式,研究高阶相互作用对群落物种共存结果的影响。研究表明:(1)高阶相互作用可以促进也可能抑制物种共存,具体共存结果取决于高阶相互作用的方向、强度和分类;(2)当全部高阶相互作用都存在,且取值为正时,物种共存位置会发生偏移,原本生态位分化下共存的区域不再共存,而在生态位重叠度较高的区域,物种可以在更大范围的适合度差异下共存;(3)扩散模式对高阶相互作用的上述调节机制有一定的影响,且无论正高阶还是负高阶,当种群趋于局部扩散时,高阶相互作用的正向及负向调节效果均有所减弱。以上结论强调了在理论模型和实际保护工作中考虑相互作用网络的重要性,有助于进一步理解物种共存机制,能够为保护生物多样性提供理论依据。     

基于支持向量机的室颤信号检测算法

目的:实现室颤信号与非室颤信号的分类,进而实现室颤信号的检测。方法:本文引入了一种基于支持向量机(Support Vec-tor Machine,SVM)和改进的越限区间算法(TCI)的新算法,其中支持向量机在处理分类和模式识别等问题中具有很大的优势。该算法采用4s的滑动窗技术,并利用改进后的越限区间算法(Threshold Crossing Interval,TCI)方法提取心电信号的特征。新算法的实现如下:在每一滑动窗内采用改进的后的绝对值阈值,计算中间2s内的平均越限间隔值。并以此TCI值作为特征参数,输入一个预先设计好的二分类支持向量机中,从而实现分类。结果:成功实现了室颤信号的检测,通过计算该方法的灵敏度、精确度、预测性和准确度且与其他方法相比较,可知此新算法总体可靠性优于其他方法。结论:该算法能够实现室颤信号的实时监测,且简单易行,易于实现,较适合实时的心电监测以及除颤仪器。     

支持向量机与神经网络的关系研究

支持向量机是一种基于统计学习理论的新颖的机器学习方法,由于其出色的学习性能,该技术已成为当前国际机器学习界的研究热点,该方法已经广泛用于解决分类和回归问题．本文将结构风险函数应用于径向基函数网络学习中,同时讨论了支持向量回归模型和径向基函数网络之间的关系．仿真实例表明所给算法提高了径向基函数网络的泛化性能．     

面向异步多时延基因调控网络建模的高阶动态贝叶斯网络模型及其结构 学习算法

目的：由基因芯片数据精确学习建模具有异步多时延表达调控关系的基因调控网络。方法：提出了一种高阶动态贝叶斯网 络模型,并给出了网络结构学习算法,该模型假定基因的调控过程为多阶马尔科夫过程,从而能够建模基因调控网络中的异步多 时延特性。结果：由酵母基因调控网络一个子网络人工生成了加入10%含噪声的表达数据用于调控网络结构学习。在75%的后验 概率下,本文提出的高阶动态贝叶斯网络模型能够正确建模实际网络中全部的异步多时延调控关系,而经典动态贝叶斯网络仅 能够正确建模实际网络中1/3的调控关系;ROC曲线对比表明在各个后验概率水平上高阶动态贝叶斯网络模型的效果均优于经 典动态贝叶斯网络。结论：本文提出的高阶动态贝叶斯网络模型能够精确学习建模具有异步多时延表达调控关系的基因调控网 络。     

面向异步多时延基因调控网络建模的高阶动态贝叶斯网络模型及其结构学习算法

目的：由基因芯片数据精确学习建模具有异步多时延表达调控关系的基因调控网络。方法：提出了一种高阶动态贝叶斯网络模型,并给出了网络结构学习算法,该模型假定基因的调控过程为多阶马尔科夫过程,从而能够建模基因调控网络中的异步多时延特性。结果：由酵母基因调控网络一个子网络人工生成了加入10%含噪声的表达数据用于调控网络结构学习。在75%的后验概率下,本文提出的高阶动态贝叶斯网络模型能够正确建模实际网络中全部的异步多时延调控关系,而经典动态贝叶斯网络仅能够正确建模实际网络中1/3的调控关系;ROC曲线对比表明在各个后验概率水平上高阶动态贝叶斯网络模型的效果均优于经典动态贝叶斯网络。结论：本文提出的高阶动态贝叶斯网络模型能够精确学习建模具有异步多时延表达调控关系的基因调控网络。     

初中生物学教学的学习性评价

以北师大版初中生物学教材中的部分单元教学活动设计为例,尝试构建有高阶思维活动的学习目标和评价要求,创设高阶认知的问题任务学习情境,开展以"思辨"为核心的学习思维发展与深化实践.实践证明,将评价活动贯穿于整个教学,并将多元化主体的有效反馈融入学习性评价,有利于促进学生的有效学习,达到教、学、评的有机统一.     

指向提升生物学实验教学品质的评价

提升对实验教学品质的评价,是追求实验教学质量与教学效果的重要方法及手段。提升生物学实验教学品质的评价主要指向:(1)发展学科核心素养即发展学生在生物学课程学习过程中逐渐发展起来的,在解决真实情境中的生物学问题时所表现出来的必备品格和关键能力;(2)发展学生高阶思维发展的高阶思维能力,主要发展批判性思维能力、创造性思维能力、问题解决能力等;(3)开发有效评价工具,优化课堂内、外实验教学的评价环境,提升实验教学品质和发展学生的实验探究能力。     

人口问题中具广义Ginzbur-Landau型方程解的渐近性和Blow-up

在文献[1][2][3]和[4]的基础上,研究人口问题中具广义Ginzbur-Landau型弥散方程（0.21）,（0.22）及更一般的非线性高阶抛物型方程（0.23）初边值问题,在古典解存在且唯一的条件下,利用文[7,8]的相应方法,深入探讨问题的广义解和古典解的渐近性及Blow-up现象,得到许多新结果。     

Modeling Higher-Order Correlations within Cortical Microcolumns

We statistically characterize the population spiking activity obtained from simultaneous recordings of neurons across all layers of a cortical microcolumn. Three types of models are compared: an Ising model which captures pairwise correlations between units, a Restricted Boltzmann Machine (RBM) which allows for modeling of higher-order correlations, and a semi-Restricted Boltzmann Machine which is a combination of Ising and RBM models. Model parameters were estimated in a fast and efficient manner using minimum probability flow, and log likelihoods were compared using annealed importance sampling. The higher-order models reveal localized activity patterns which reflect the laminar organization of neurons within a cortical column. The higher-order models also outperformed the Ising model in log-likelihood: On populations of 20 cells, the RBM had 10% higher log-likelihood (relative to an independent model) than a pairwise model, increasing to 45% gain in a larger network with 100 spatiotemporal elements, consisting of 10 neurons over 10 time steps. We further removed the need to model stimulus-induced correlations by incorporating a peri-stimulus time histogram term, in which case the higher order models continued to perform best. These results demonstrate the importance of higher-order interactions to describe the structure of correlated activity in cortical networks. Boltzmann Machines with hidden units provide a succinct and effective way to capture these dependencies without increasing the difficulty of model estimation and evaluation.     

Active Semi-Supervised Learning Method with Hybrid Deep Belief Networks

In this paper, we develop a novel semi-supervised learning algorithm called active hybrid deep belief networks (AHD), to address the semi-supervised sentiment classification problem with deep learning. First, we construct the previous several hidden layers using restricted Boltzmann machines (RBM), which can reduce the dimension and abstract the information of the reviews quickly. Second, we construct the following hidden layers using convolutional restricted Boltzmann machines (CRBM), which can abstract the information of reviews effectively. Third, the constructed deep architecture is fine-tuned by gradient-descent based supervised learning with an exponential loss function. Finally, active learning method is combined based on the proposed deep architecture. We did several experiments on five sentiment classification datasets, and show that AHD is competitive with previous semi-supervised learning algorithm. Experiments are also conducted to verify the effectiveness of our proposed method with different number of labeled reviews and unlabeled reviews respectively.     

Multi-scale detection of rate changes in spike trains with weak dependencies

The statistical analysis of neuronal spike trains by models of point processes often relies on the assumption of constant process parameters. However, it is a well-known problem that the parameters of empirical spike trains can be highly variable, such as for example the firing rate. In order to test the null hypothesis of a constant rate and to estimate the change points, a Multiple Filter Test (MFT) and a corresponding algorithm (MFA) have been proposed that can be applied under the assumption of independent inter spike intervals (ISIs). As empirical spike trains often show weak dependencies in the correlation structure of ISIs, we extend the MFT here to point processes associated with short range dependencies. By specifically estimating serial dependencies in the test statistic, we show that the new MFT can be applied to a variety of empirical firing patterns, including positive and negative serial correlations as well as tonic and bursty firing. The new MFT is applied to a data set of empirical spike trains with serial correlations, and simulations show improved performance against methods that assume independence. In case of positive correlations, our new MFT is necessary to reduce the number of false positives, which can be highly enhanced when falsely assuming independence. For the frequent case of negative correlations, the new MFT shows an improved detection probability of change points and thus, also a higher potential of signal extraction from noisy spike trains.     

采用Reichardt运动检测器和Boltzmann Machine神经网络的运动计算

建立了一个探讨灵长类视皮层从Ｖ１区到ＭＴ区的运动信息加工原理的计算模型,这个过程的突出特征是视觉运动信息经过了从局部检测进步到整体感知。模型的第一层由用于抽提运动模式的局部速度以及结构性质的Ｒｅｉｃｈａｒｄｔ运动检测器组成,进一步的加工是通过Ｂｏｌｔｚｍａｎｎ Ｍａｃｈｉｎｅ神经网络来实现的。这种网络的学习算法具有局部更新的显著性质,在学习阶段,网络不断地修改联结权重以形成对于记录在网络的显单元上     

Instabilities and oscillation in the deterministic Boltzmann machine

Simulations indicate that the deterministic Boltzmann machine, unlike the stochastic Boltzmann machine from which it is derived, exhibits unstable behavior during contrastive Hebbian learning of nonlinear problems, including oscillation in the learning algorithm and extreme sensitivity to small weight perturbations. Although careful choice of the initial weight magnitudes, the learning rate, and the annealing schedule will produce convergence in most cases, the stability of the resulting solution depends on the parameters in a complex and generally indiscernible way. We show that this unstable behavior is the result of over parameterization (excessive freedom in the weights), which leads to continuous rather than isolated optimal weight solution sets. This allows the weights to drift without correction by the learning algorithm until the free energy landscape changes in such a way that the settling procedure employed finds a different minimum of the free energy function than it did previously and a gross output error occurs. Because all the weight sets in a continuous optimal solution set produce exactly the same network outputs, we define reliability, a measure of the robustness of the network, as a new performance criterion.     

Acceleration of flowering by overexpression of MFT (MOTHER OF FT AND TFL1)

MFT (MOTHER OF FT AND TFL1) is a member of a gene family that includes two important regulators, FT (FLOWERING LOCUS T) and TFL1 (TERMINAL FLOWER 1), in determination of flowering time in Arabidopsis. Although the functions of FT and TFL1 are assigned in the family, the roles of other members are largely unknown. Especially the sequence of MFT is homologous to both FT and TFL1, which act as a floral promoter and an inhibitor, respectively, making it difficult to predict the role of MFT. We performed genetic analyses of MFT to understand its role in floral development. Constitutive expression of MFT led to slightly early flowering under long days. However, a T-DNA insertion allele of MFT did not show obvious phenotype. Further genetic analyses with the loss-of-function alleles of FT, TFL1, and ATC (Arabidopsis Thaliana CENTRORADIALIS homologue) showed that a decrease of MFT activity did not enhance the phenotypes of the single mutants. Taken together, we suggest that MFT functions as a floral inducer and that it may act redundantly in determination of flowering time in Arabidopsis.     

Kernel hierarchical gene clustering from microarray expression data

MOTIVATION: Unsupervised analysis of microarray gene expression data attempts to find biologically significant patterns within a given collection of expression measurements. For example, hierarchical clustering can be applied to expression profiles of genes across multiple experiments, identifying groups of genes that share similar expression profiles. Previous work using the support vector machine supervised learning algorithm with microarray data suggests that higher-order features, such as pairwise and tertiary correlations across multiple experiments, may provide significant benefit in learning to recognize classes of co-expressed genes. RESULTS: We describe a generalization of the hierarchical clustering algorithm that efficiently incorporates these higher-order features by using a kernel function to map the data into a high-dimensional feature space. We then evaluate the utility of the kernel hierarchical clustering algorithm using both internal and external validation. The experiments demonstrate that the kernel representation itself is insufficient to provide improved clustering performance. We conclude that mapping gene expression data into a high-dimensional feature space is only a good idea when combined with a learning algorithm, such as the support vector machine that does not suffer from the curse of dimensionality. AVAILABILITY: Supplementary data at www.cs.columbia.edu/compbio/hiclust. Software source code available by request.     

环境微生物研究中机器学习算法及应用

微生物在环境中无处不在，它们不仅是生物地球化学循环和环境演化的关键参与者，也在环境监测、生态治理和保护中发挥着重要作用。随着高通量技术的发展，大量微生物数据产生，运用机器学习对环境微生物大数据进行建模和分析，在微生物标志物识别、污染物预测和环境质量预测等领域的科学研究和社会应用方面均具有重要意义。机器学习可分为监督学习和无监督学习2大类。在微生物组学研究当中，无监督学习通过聚类、降维等方法高效地学习输入数据的特征，进而对微生物数据进行整合和归类。监督学习运用有特征和标记的微生物数据集训练模型，在面对只有特征没有标记的数据时可以判断出标记，从而实现对新数据的分类、识别和预测。然而，复杂的机器学习算法通常以牺牲可解释性为代价来重点关注模型预测的准确性。机器学习模型通常可以看作预测特定结果的“黑匣子”，即对模型如何得出预测所知甚少。为了将机器学习更多地运用于微生物组学研究、提高我们提取有价值的微生物信息的能力，深入了解机器学习算法、提高模型的可解释性尤为重要。本文主要介绍在环境微生物领域常用的机器学习算法和基于微生物组数据的机器学习模型的构建步骤，包括特征选择、算法选择、模型构建和评估等，并对各种机器学习模型在环境微生物领域的应用进行综述，深入探究微生物组与周围环境之间的关联，探讨提高模型可解释性的方法，并为未来环境监测、环境健康预测提供科学参考。     

Visuospatial Sequence Learning without Seeing

Background

The ability to detect and integrate associations between unrelated items that are close in space and time is a key feature of human learning and memory. Learning sequential associations between non-adjacent visual stimuli (higher-order visuospatial dependencies) can occur either with or without awareness (explicit vs. implicit learning) of the products of learning. Existing behavioural and neurocognitive studies of explicit and implicit sequence learning, however, are based on conscious access to the sequence of target locations and, typically, on conditions where the locations for orienting, or motor, responses coincide with the locations of the target sequence.

Methodology/Principal Findings

Dichoptic stimuli were presented on a novel sequence learning task using a mirror stereoscope to mask the eye-of-origin of visual input from conscious awareness. We demonstrate that conscious access to the sequence of target locations and responses that coincide with structure of the target sequence are dispensable features when learning higher-order visuospatial associations. Sequence knowledge was expressed in the ability of participants to identify the trained higher-order visuospatial sequence on a recognition test, even though the trained and untrained recognition sequences were identical when viewed at a conscious binocular level, and differed only at the level of the masked sequential associations.

Conclusions/Significance

These results demonstrate that unconscious processing can support perceptual learning of higher-order sequential associations through interocular integration of retinotopic-based codes stemming from monocular eye-of-origin information. Furthermore, unlike other forms of perceptual associative learning, visuospatial attention did not need to be directed to the locations of the target sequence. More generally, the results pose a challenge to neural models of learning to account for a previously unknown capacity of the human visual system to support the detection, learning and recognition of higher-order sequential associations under conditions where observers are unable to see the target sequence or perform responses that coincide with structure of the target sequence.