The evolutionary language game.

We explore how evolutionary game dynamics have to be modified to accomodate a mathematical framework for the evolution of language. In particular, we are interested in the evolution of vocabulary, that is associations between signals and objects. We assume that successful communication contributes to biological fitness: individuals who communicate well leave more offspring. Children inherit from their parents a strategy for language learning (a language acquisition device). We consider three mechanisms whereby language is passed from one generation to the next: (i) parental learning: children learn the language of their parents; (ii) role model learning: children learn the language of individuals with a high payoff; and (iii) random learning: children learn the language of randomly chosen individuals. We show that parental and role model learning outperform random learning. Then we introduce mistakes in language learning and study how this process changes language over time. Mistakes increase the overall efficacy of parental and role model learning: in a world with errors evolutionary adaptation is more efficient. Our model also provides a simple explanation why homonomy is common while synonymy is rare.     

Phenotype Classification of Zebrafish Embryos by Supervised Learning

Zebrafish is increasingly used to assess biological properties of chemical substances and thus is becoming a specific tool for toxicological and pharmacological studies. The effects of chemical substances on embryo survival and development are generally evaluated manually through microscopic observation by an expert and documented by several typical photographs. Here, we present a methodology to automatically classify brightfield images of wildtype zebrafish embryos according to their defects by using an image analysis approach based on supervised machine learning. We show that, compared to manual classification, automatic classification results in 90 to 100% agreement with consensus voting of biological experts in nine out of eleven considered defects in 3 days old zebrafish larvae. Automation of the analysis and classification of zebrafish embryo pictures reduces the workload and time required for the biological expert and increases the reproducibility and objectivity of this classification.     

Compressed learning and its applications to subcellular localization

One of the main challenges faced by biological applications is to predict protein subcellular localization in automatic fashion accurately. To achieve this in these applications, a wide variety of machine learning methods have been proposed in recent years. Most of them focus on finding the optimal classification scheme and less of them take the simplifying the complexity of biological systems into account. Traditionally, such bio-data are analyzed by first performing a feature selection before classification. Motivated by CS (Compressed Sensing) theory, we propose the methodology which performs compressed learning with a sparseness criterion such that feature selection and dimension reduction are merged into one analysis. The proposed methodology decreases the complexity of biological system, while increases protein subcellular localization accuracy. Experimental results are quite encouraging, indicating that the aforementioned sparse methods are quite promising in dealing with complicated biological problems, such as predicting the subcellular localization of Gram-negative bacterial proteins.     

Dendritic normalisation improves learning in sparsely connected artificial neural networks

Artificial neural networks, taking inspiration from biological neurons, have become an invaluable tool for machine learning applications. Recent studies have developed techniques to effectively tune the connectivity of sparsely-connected artificial neural networks, which have the potential to be more computationally efficient than their fully-connected counterparts and more closely resemble the architectures of biological systems. We here present a normalisation, based on the biophysical behaviour of neuronal dendrites receiving distributed synaptic inputs, that divides the weight of an artificial neuron’s afferent contacts by their number. We apply this dendritic normalisation to various sparsely-connected feedforward network architectures, as well as simple recurrent and self-organised networks with spatially extended units. The learning performance is significantly increased, providing an improvement over other widely-used normalisations in sparse networks. The results are two-fold, being both a practical advance in machine learning and an insight into how the structure of neuronal dendritic arbours may contribute to computation.     

Inferring adaptive regulation thresholds and association rules from gene expression data through combinatorial optimization learning

There is a need to design computational methods to support the prediction of gene regulatory networks. Such models should offer both biologically-meaningful and computationally-accurate predictions, which in combination with other techniques may improve large-scale, integrative studies. This paper presents a new machine learning method for the prediction of putative regulatory associations from expression data, which exhibit properties never or only partially addressed by other techniques recently published. The method was tested on a Saccharomyces cerevisiae gene expression dataset. The results were statistically validated and compared with the relationships inferred by two machine learning approaches to gene regulatory network prediction. Furthermore, the resulting predictions were assessed using domain knowledge. The proposed algorithm may be able to accurately predict relevant biological associations between genes. One of the most relevant features of this new method is the prediction of adaptive regulation thresholds for the discretization of gene expression values, which is required prior to the rule association learning process. Moreover, an important advantage consists of its low computational cost to infer association rules. The proposed system may significantly support exploratory, large-scale studies of automated identification of potentially-relevant gene expression associations.     

Candidate gene analysis for quantitative traits using the transmission disequilibrium test: the example of the melanocortin 4-receptor in pigs

Population-wide associations between loci due to linkage disequilibrium can be used to map quantitative trait loci (QTL) with high resolution. However, spurious associations between markers and QTL can also arise as a consequence of population stratification. Statistical methods that cannot differentiate between loci associations due to linkage disequilibria from those caused in other ways can render false-positive results. The transmission-disequilibrium test (TDT) is a robust test for detecting QTL. The TDT exploits within-family associations that are not affected by population stratification. However, some TDTs are formulated in a rigid form, with reduced potential applications. In this study we generalize TDT using mixed linear models to allow greater statistical flexibility. Allelic effects are estimated with two independent parameters: one exploiting the robust within-family information and the other the potentially biased between-family information. A significant difference between these two parameters can be used as evidence for spurious association. This methodology was then used to test the effects of the fourth melanocortin receptor (MC4R) on production traits in the pig. The new analyses supported the previously reported results; i.e., the studied polymorphism is either causal or in very strong linkage disequilibrium with the causal mutation, and provided no evidence for spurious association.     

Oral health and frailty in the medieval English cemetery of St Mary Graces

The analysis of oral pathologies is routinely a part of bioarcheological and paleopathological investigations. Oral health, while certainly interesting by itself, is also potentially informative about general or systemic health. Numerous studies within modern populations have shown associations between oral pathologies and other diseases, such as cardiovascular disease, certain types of cancer, and pulmonary infections. This article addresses the question of how oral health was associated with general health in past populations by examining the relationship between two oral pathologies (periodontal disease and dental caries) and the risk of mortality in a cemetery sample from medieval England. The effects of periodontitis and dental caries on risk of death were assessed using a sample of 190 individuals from the St Mary Graces cemetery, London, dating to ～AD 1350–1538. The results suggest that the oral pathologies are associated with elevated risks of mortality in the St Mary Graces cemetery such that individuals with periodontitis and dental caries were more likely to die than their peers without such pathologies. The results shown here suggest that these oral pathologies can be used as informative indicators of general health in past populations. Am J Phys Anthropol, 2010. © 2009 Wiley‐Liss, Inc.     

Application of Artificial Intelligence/Machine Vision & Learning for the Development of a Live Single-cell Phenotypic Biomarker Test to Predict Prostate Cancer Tumor Aggressiveness

To assess the usefulness and applications of machine vision (MV) and machine learning (ML) techniques that have been used to develop a single cell-based phenotypic (live and fixed biomarkers) platform that correlates with tumor biological aggressiveness and risk stratification, 100 fresh prostate samples were acquired, and areas of prostate cancer were determined by post-surgery pathology reports logged by an independent pathologist. The prostate samples were dissociated into single-cell suspensions in the presence of an extracellular matrix formulation. These samples were analyzed via live-cell microscopy. Dynamic and fixed phenotypic biomarkers per cell were quantified using objective MV software and ML algorithms. The predictive nature of the ML algorithms was developed in two stages. First, random forest (RF) algorithms were developed using 70% of the samples. The developed algorithms were then tested for their predictive performance using the blinded test dataset that contained 30% of the samples in the second stage. Based on the ROC (receiver operating characteristic) curve analysis, thresholds were set to maximize both sensitivity and specificity. We determined the sensitivity and specificity of the assay by comparing the algorithm-generated predictions with adverse pathologic features in the radical prostatectomy (RP) specimens. Using MV and ML algorithms, the biomarkers predictive of adverse pathology at RP were ranked and a prostate cancer patient risk stratification test was developed that distinguishes patients based on surgical adverse pathology features. The ability to identify and track large numbers of individual cells over the length of the microscopy experimental monitoring cycles, in an automated way, created a large biomarker dataset of primary biomarkers. This biomarker dataset was then interrogated with ML algorithms used to correlate with post-surgical adverse pathology findings. Algorithms were generated that predicted adverse pathology with >0.85 sensitivity and specificity and an AUC (area under the curve) of >0.85. Phenotypic biomarkers provide cellular and molecular details that are informative for predicting post-surgical adverse pathologies when considering tumor biopsy samples. Artificial intelligence ML-based approaches for cancer risk stratification are emerging as important and powerful tools to compliment current measures of risk stratification. These techniques have capabilities to address tumor heterogeneity and the molecular complexity of prostate cancer. Specifically, the phenotypic test is a novel example of leveraging biomarkers and advances in MV and ML for developing a powerful prognostic and risk-stratification tool for prostate cancer patients.     

System-wide analysis of hepatotoxicological responses: tissomics is key.

BACKGROUND: Combining diverse data streams across different levels of biological observation, such as molecular, cellular, and clinical chemistry responses, support a system-wide diagnostic approach. Recent progress in slide-based cytometry contributes to the development of tissomics, a high-throughput and high-content phenotyping methodology that provides data-rich profiles of cellular heterogeneity in tissues enabling correlative statistical treatments over multiple scales of biological hierarchies. METHODS: Phenotypical data are covariants that can be used as biomarkers to identify relevant candidate genes by associating initiating molecular events with phenotypical changes and adverse outcomes. We introduce a procedure of combined statistical and analytical tools to identify and visualize such associations for nonpooled entities. The new utility is applied to a time-controlled, low-dose toxicological study including a control and two xenobiotic compounds. RESULTS: An integrated analysis identified specific molecular and phenotypical biomarkers, which support the classification of animals in the absence of any visual indicators from pathology readings. DISCUSSION: The introduction of controlled perturbations to tissues provides a prototypical setting to develop a sensitive, systems-based analysis methodology suitable for a broader range of biomedical applications.     

Perspectives,potentials and trends of ex vivo and in vivo optical molecular pathology

It is pivotal for medical applications, such as noninvasive histopathologic characterization of tissue, to realize label‐free and molecule‐specific representation of morphologic and biochemical composition in real‐time with subcellular spatial resolution. This unmet clinical need requires new approaches for rapid and reliable real‐time assessment of pathologies to complement established diagnostic tools. Photonic imaging combined with digitalization offers the potential to provide the clinician the requested information both under in vivo and ex vivo conditions. This report summarizes photonic approaches and their use in combination with image processing, machine learning and augmented virtual reality that might solve current challenges in modern medicine. Details are given for pathology, intraoperative diagnosis in head and neck cancer and endoscopic diagnosis in gastroenterology.      

Machine learning approaches applied in spinal pain research

The purpose of this narrative review is to provide a critical reflection of how analytical machine learning approaches could provide the platform to harness variability of patient presentation to enhance clinical prediction. The review includes a summary of current knowledge on the physiological adaptations present in people with spinal pain. We discuss how contemporary evidence highlights the importance of not relying on single features when characterizing patients given the variability of physiological adaptations present in people with spinal pain. The advantages and disadvantages of current analytical strategies in contemporary basic science and epidemiological research are reviewed and we consider how analytical machine learning approaches could provide the platform to harness the variability of patient presentations to enhance clinical prediction of pain persistence or recurrence. We propose that machine learning techniques can be leveraged to translate a potentially heterogeneous set of variables into clinically useful information with the potential to enhance patient management.     

The different roles of social learning in vocal communication

While vocal learning has been studied extensively in birds and mammals, little effort has been made to define what exactly constitutes vocal learning and to classify the forms that it may take. We present such a theoretical framework for the study of social learning in vocal communication. We define different forms of social learning that affect communication and discuss the required methodology to show each one. We distinguish between contextual and production learning in animal communication. Contextual learning affects the behavioural context or serial position of a signal. It can affect both usage and comprehension. Production learning refers to instances where the signals themselves are modified in form as a result of experience with those of other individuals. Vocal learning is defined as production learning in the vocal domain. It can affect one or more of three systems: the respiratory, phonatory and filter systems. Each involves a different level of control over the sound production apparatus. We hypothesize that contextual learning and respiratory production learning preceded the evolution of phonatory and filter production learning. Each form of learning potentially increases the complexity of a communication system. We also found that unexpected genetic or environmental factors can have considerable effects on vocal behaviour in birds and mammals and are often more likely to cause changes or differences in vocalizations than investigators may assume. Finally, we discuss how production learning is used in innovation and invention, and present important future research questions. Copyright 2000 The Association for the Study of Animal Behaviour.     

A model of associative learning in the mushroom body

The mushroom body is a prominent invertebrate neuropil strongly associated with learning and memory. We built a high-level computational model of this structure using simplified but realistic models of neurons and synapses, and developed a learning rule based on activity dependent pre-synaptic facilitation. We show that our model, which is consistent with mushroom body Drosophila data and incorporates Aplysia learning, is able to both acquire and later recall CS-US associations. We demonstrate that a highly divergent input connectivity to the mushroom body and strong periodic inhibition both serve to improve overall learning performance. We also examine the problem of how synaptic conductance, driven by successive training events, obtains a value appropriate for the stimulus being learnt. We employ two feedback mechanisms: one stabilises strength at an initial level appropriate for an association; another prevents strength increase for established associations.     

The evolution of sperm length in moths

Sperm form and function remain poorly understood despite being of fundamental biological importance. An instructive approach has been to examine evolutionary associations across comparable taxa between sperm characters and other, potentially selective reproductive traits. We adopt this approach here in a comparative study examining how sperm lengths are associated with male and female reproductive characters across moths. Primary data have revealed Lepidoptera to be an ideal order for examination: there is profound variation in the dimensions (but not organization) of the reproductive traits between closely related species which all share a monophyletic ancestry, for example, eupyrene sperm length varies from 110 to 12,675 microm. Eupyrene (normal fertilizing) and apyrene (anucleate and non-fertile) sperm lengths are positively correlated across taxa and both sperm types show positive associations with mating pattern (as measured by the residual testis size). At fertilization, eupyrene sperm must migrate down the often elongated female spermathecal duct from storage to unite with the ovum. Across taxa, the elongation of this duct is associated with increased eupyrene sperm length, suggesting a positive female influence on sperm size since longer, more powerful sperm may be selected to migrate and/or compete successfully down greater ductal lengths. Apyrene sperm length is not associated with female reproductive tract dimensions. However, we found a positive relationship between the residual testis volume and spermathecal volume, suggesting coevolution between male investment in spermatogenesis and the extent of the female sperm storage capacity. Within males, there is a positive association between the two organs which form the ejaculate-containing spermatophore: the testes and the accessory gland. The 'trade-up' in investment to these components is discussed in relation to paternal investment and mating patterns.     

Diagnosis and follow-up of cystinuria: Use of proton magnetic resonance spectroscopy

Summary. Proton Nuclear Magnetic Resonance (NMR) Spectroscopy of urine (as well as of other biological fluids) is a very powerful technique enabling multi-component analysis useful in both diagnosis and follow-up of a wide range of inherited metabolic diseases. Among these pathologies, cystinuria is characterised by accumulation in urine of four dibasic amino acids, namely lysine, arginine, ornithine and cystine; the last one, being only slightly water soluble, generates urolithiasis. The mentioned aminoacids can be detected in the urine NMR spectrum of cystinuric patients, the most abundant being the lysine (5 mM and over are often detected), whose typical signals become very high; arginine and ornithine are also usually detectable, although pathologic concentrations are lower (usually below 2 mM). The proposed NMR technique is also suitable in monitoring the therapy with α-mercaptopropionylglycine (MPG), providing quantitation of several metabolites of interest in the follow-up of the pathology, like cystine, creatinine and citrate. Received May 9, 1999; Accepted September 26, 1999     

机器学习在MALDI-TOF MS鉴定微生物中的应用

基质辅助激光解吸/电离飞行时间质谱(matrix-assisted laser desorption/ionization time-of-flight mass spectrometry,MALDI-TOF MS)是一种新兴的高通量技术,已广泛应用于临床微生物、食品微生物和水产微生物的快速鉴定。如何进一步提高MALDI-TOF MS在微生物鉴定中的分辨率是该技术当前面临的一大挑战。为了高效处理大量高维微生物MALDI-TOF MS数据,各种机器学习算法得到了应用。本文综述了机器学习在微生物MALDI-TOFMS鉴定中的应用。首先,本文在介绍机器学习在微生物MALDI-TOF MS分类中的工作流程后,进一步对MALDI-TOF MS的数据特征、MALDI-TOF MS数据库、数据的预处理和模型的性能评估进行了描述。然后讨论了典型的机器学习分类算法和集成学习算法的应用。简单的机器学习算法很难满足微生物MALDI-TOF MS分类的高分辨率的需求,而组合不同机器学习算法和集成学习算法可以获得更好的微生物分类性能。在MALDI-TOF MS数据的预处理方面,小波算法和遗传算法的应用最广,它们...     

Interpreting human genetic variation with in vivo zebrafish assays

Rapid advances and cost erosion in exome and genome analysis of patients with both rare and common genetic disorders have accelerated gene discovery and illuminated fundamental biological mechanisms. The thrill of discovery has been accompanied, however, with the sobering appreciation that human genomes are burdened with a large number of rare and ultra rare variants, thereby posing a significant challenge in dissecting both the effect of such alleles on protein function and also the biological relevance of these events to patient pathology. In an effort to develop model systems that are able to generate surrogates of human pathologies, a powerful suite of tools have been developed in zebrafish, taking advantage of the relatively small (compared to invertebrate models) evolutionary distance of that genome to humans, the orthology of several organs and signaling processes, and the suitability of this organism for medium and high throughput phenotypic screening. Here we will review the use of this model organism in dissecting human genetic disorders; we will highlight how diverse strategies have informed disease causality and genetic architecture; and we will discuss relative strengths and limitations of these approaches in the context of medical genome sequencing. This article is part of a Special Issue entitled: From Genome to Function.