Searching for novel gene functions in yeast : identification of thousands of novel molecular interactions by protein-fragment complementation assay followed by automated gene function prediction and high-throughput lipidomics

La compréhension de processus biologiques complexes requiert des approches expérimentales et informatiques sophistiquées. Les récents progrès dans le domaine des stratégies génomiques fonctionnelles mettent dorénavant à notre disposition de puissants outils de collecte de données sur l’interconnectivité des gènes, des protéines et des petites molécules, dans le but d’étudier les principes organisationnels de leurs réseaux cellulaires. L’intégration de ces connaissances au sein d’un cadre de référence en biologie systémique permettrait la prédiction de nouvelles fonctions de gènes qui demeurent non caractérisées à ce jour. Afin de réaliser de telles prédictions à l’échelle génomique chez la levure Saccharomyces cerevisiae, nous avons développé une stratégie innovatrice qui combine le criblage interactomique à haut débit des interactions protéines-protéines, la prédiction de la fonction des gènes in silico ainsi que la validation de ces prédictions avec la lipidomique à haut débit. D’abord, nous avons exécuté un dépistage à grande échelle des interactions protéines-protéines à l’aide de la complémentation de fragments protéiques. Cette méthode a permis de déceler des interactions in vivo entre les protéines exprimées par leurs promoteurs naturels. De plus, aucun biais lié aux interactions des membranes n’a pu être mis en évidence avec cette méthode, comparativement aux autres techniques existantes qui décèlent les interactions protéines-protéines. Conséquemment, nous avons découvert plusieurs nouvelles interactions et nous avons augmenté la couverture d’un interactome d’homéostasie lipidique dont la compréhension demeure encore incomplète à ce jour. Par la suite, nous avons appliqué un algorithme d’apprentissage afin d’identifier huit gènes non caractérisés ayant un rôle potentiel dans le métabolisme des lipides. Finalement, nous avons étudié si ces gènes et un groupe de régulateurs transcriptionnels distincts, non préalablement impliqués avec les lipides, avaient un rôle dans l’homéostasie des lipides. Dans ce but, nous avons analysé les lipidomes des délétions mutantes de gènes sélectionnés. Afin d’examiner une grande quantité de souches, nous avons développé une plateforme à haut débit pour le criblage lipidomique à contenu élevé des bibliothèques de levures mutantes. Cette plateforme consiste en la spectrométrie de masse à haute resolution Orbitrap et en un cadre de traitement des données dédié et supportant le phénotypage des lipides de centaines de mutations de Saccharomyces cerevisiae. Les méthodes expérimentales en lipidomiques ont confirmé les prédictions fonctionnelles en démontrant certaines différences au sein des phénotypes métaboliques lipidiques des délétions mutantes ayant une absence des gènes YBR141C et YJR015W, connus pour leur implication dans le métabolisme des lipides. Une altération du phénotype lipidique a également été observé pour une délétion mutante du facteur de transcription KAR4 qui n’avait pas été auparavant lié au métabolisme lipidique. Tous ces résultats démontrent qu’un processus qui intègre l’acquisition de nouvelles interactions moléculaires, la prédiction informatique des fonctions des gènes et une plateforme lipidomique innovatrice à haut débit , constitue un ajout important aux méthodologies existantes en biologie systémique. Les développements en méthodologies génomiques fonctionnelles et en technologies lipidomiques fournissent donc de nouveaux moyens pour étudier les réseaux biologiques des eucaryotes supérieurs, incluant les mammifères. Par conséquent, le stratégie présenté ici détient un potentiel d’application au sein d’organismes plus complexes.

Distributed conditional computation

L'objectif de cette thèse est de présenter différentes applications du programme de recherche de calcul conditionnel distribué. On espère que ces applications, ainsi que la théorie présentée ici, mènera à une solution générale du problème d'intelligence artificielle, en particulier en ce qui a trait à la nécessité d'efficience. La vision du calcul conditionnel distribué consiste à accélérer l'évaluation et l'entraînement de modèles profonds, ce qui est très différent de l'objectif usuel d'améliorer sa capacité de généralisation et d'optimisation. Le travail présenté ici a des liens étroits avec les modèles de type mélange d'experts. Dans le chapitre 2, nous présentons un nouvel algorithme d'apprentissage profond qui utilise une forme simple d'apprentissage par renforcement sur un modèle d'arbre de décisions à base de réseau de neurones. Nous démontrons la nécessité d'une contrainte d'équilibre pour maintenir la distribution d'exemples aux experts uniforme et empêcher les monopoles. Pour rendre le calcul efficient, l'entrainement et l'évaluation sont contraints à être éparse en utilisant un routeur échantillonnant des experts d'une distribution multinomiale étant donné un exemple. Dans le chapitre 3, nous présentons un nouveau modèle profond constitué d'une représentation éparse divisée en segments d'experts. Un modèle de langue à base de réseau de neurones est construit à partir des transformations éparses entre ces segments. L'opération éparse par bloc est implémentée pour utilisation sur des cartes graphiques. Sa vitesse est comparée à deux opérations denses du même calibre pour démontrer le gain réel de calcul qui peut être obtenu. Un modèle profond utilisant des opérations éparses contrôlées par un routeur distinct des experts est entraîné sur un ensemble de données d'un milliard de mots. Un nouvel algorithme de partitionnement de données est appliqué sur un ensemble de mots pour hiérarchiser la couche de sortie d'un modèle de langage, la rendant ainsi beaucoup plus efficiente. Le travail présenté dans cette thèse est au centre de la vision de calcul conditionnel distribué émis par Yoshua Bengio. Elle tente d'appliquer la recherche dans le domaine des mélanges d'experts aux modèles profonds pour améliorer leur vitesse ainsi que leur capacité d'optimisation. Nous croyons que la théorie et les expériences de cette thèse sont une étape importante sur la voie du calcul conditionnel distribué car elle cadre bien le problème, surtout en ce qui concerne la compétitivité des systèmes d'experts.

AN ARTIFICIAL NEURAL NETWORK FOR SONAR TARGET DETECTION

Neural Network has emerged as the topic of the day. The spectrum of its application is as wide as from ECG noise filtering to seismic data analysis and from elementary particle detection to electronic music composition. The focal point of the proposed work is an application of a massively parallel connectionist model network for detection of a sonar target. This task is segmented into: (i) generation of training patterns from sea noise that contains radiated noise of a target, for teaching the network;(ii) selection of suitable network topology and learning algorithm and (iii) training of the network and its subsequent testing where the network detects, in unknown patterns applied to it, the presence of the features it has already learned in. A three-layer perceptron using backpropagation learning is initially subjected to a recursive training with example patterns (derived from sea ambient noise with and without the radiated noise of a target). On every presentation, the error in the output of the network is propagated back and the weights and the bias associated with each neuron in the network are modified in proportion to this error measure. During this iterative process, the network converges and extracts the target features which get encoded into its generalized weights and biases.In every unknown pattern that the converged network subsequently confronts with, it searches for the features already learned and outputs an indication for their presence or absence. This capability for target detection is exhibited by the response of the network to various test patterns presented to it.Three network topologies are tried with two variants of backpropagation learning and a grading of the performance of each combination is subsequently made.

Understanding Subsystems in Biology through Dimensionality Reduction, Graph Partitioning and Analytical Modeling

Biological systems exhibit rich and complex behavior through the orchestrated interplay of a large array of components. It is hypothesized that separable subsystems with some degree of functional autonomy exist; deciphering their independent behavior and functionality would greatly facilitate understanding the system as a whole. Discovering and analyzing such subsystems are hence pivotal problems in the quest to gain a quantitative understanding of complex biological systems. In this work, using approaches from machine learning, physics and graph theory, methods for the identification and analysis of such subsystems were developed. A novel methodology, based on a recent machine learning algorithm known as non-negative matrix factorization (NMF), was developed to discover such subsystems in a set of large-scale gene expression data. This set of subsystems was then used to predict functional relationships between genes, and this approach was shown to score significantly higher than conventional methods when benchmarking them against existing databases. Moreover, a mathematical treatment was developed to treat simple network subsystems based only on their topology (independent of particular parameter values). Application to a problem of experimental interest demonstrated the need for extentions to the conventional model to fully explain the experimental data. Finally, the notion of a subsystem was evaluated from a topological perspective. A number of different protein networks were examined to analyze their topological properties with respect to separability, seeking to find separable subsystems. These networks were shown to exhibit separability in a nonintuitive fashion, while the separable subsystems were of strong biological significance. It was demonstrated that the separability property found was not due to incomplete or biased data, but is likely to reflect biological structure.

Factorial Hidden Markov Models

We present a framework for learning in hidden Markov models with distributed state representations. Within this framework, we derive a learning algorithm based on the Expectation--Maximization (EM) procedure for maximum likelihood estimation. Analogous to the standard Baum-Welch update rules, the M-step of our algorithm is exact and can be solved analytically. However, due to the combinatorial nature of the hidden state representation, the exact E-step is intractable. A simple and tractable mean field approximation is derived. Empirical results on a set of problems suggest that both the mean field approximation and Gibbs sampling are viable alternatives to the computationally expensive exact algorithm.

The Unsupervised Acquisition of a Lexicon from Continuous Speech

We present an unsupervised learning algorithm that acquires a natural-language lexicon from raw speech. The algorithm is based on the optimal encoding of symbol sequences in an MDL framework, and uses a hierarchical representation of language that overcomes many of the problems that have stymied previous grammar-induction procedures. The forward mapping from symbol sequences to the speech stream is modeled using features based on articulatory gestures. We present results on the acquisition of lexicons and language models from raw speech, text, and phonetic transcripts, and demonstrate that our algorithm compares very favorably to other reported results with respect to segmentation performance and statistical efficiency.

On the Convergence of Stochastic Iterative Dynamic Programming Algorithms

Recent developments in the area of reinforcement learning have yielded a number of new algorithms for the prediction and control of Markovian environments. These algorithms, including the TD(lambda) algorithm of Sutton (1988) and the Q-learning algorithm of Watkins (1989), can be motivated heuristically as approximations to dynamic programming (DP). In this paper we provide a rigorous proof of convergence of these DP-based learning algorithms by relating them to the powerful techniques of stochastic approximation theory via a new convergence theorem. The theorem establishes a general class of convergent algorithms to which both TD(lambda) and Q-learning belong.

Support Vector Machines: Training and Applications

The Support Vector Machine (SVM) is a new and very promising classification technique developed by Vapnik and his group at AT&T Bell Labs. This new learning algorithm can be seen as an alternative training technique for Polynomial, Radial Basis Function and Multi-Layer Perceptron classifiers. An interesting property of this approach is that it is an approximate implementation of the Structural Risk Minimization (SRM) induction principle. The derivation of Support Vector Machines, its relationship with SRM, and its geometrical insight, are discussed in this paper. Training a SVM is equivalent to solve a quadratic programming problem with linear and box constraints in a number of variables equal to the number of data points. When the number of data points exceeds few thousands the problem is very challenging, because the quadratic form is completely dense, so the memory needed to store the problem grows with the square of the number of data points. Therefore, training problems arising in some real applications with large data sets are impossible to load into memory, and cannot be solved using standard non-linear constrained optimization algorithms. We present a decomposition algorithm that can be used to train SVM's over large data sets. The main idea behind the decomposition is the iterative solution of sub-problems and the evaluation of, and also establish the stopping criteria for the algorithm. We present previous approaches, as well as results and important details of our implementation of the algorithm using a second-order variant of the Reduced Gradient Method as the solver of the sub-problems. As an application of SVM's, we present preliminary results we obtained applying SVM to the problem of detecting frontal human faces in real images.

Object and Scene Classification: what does a Supervised Approach Provide us?

Given a set of images of scenes containing different object categories (e.g. grass, roads) our objective is to discover these objects in each image, and to use this object occurrences to perform a scene classification (e.g. beach scene, mountain scene). We achieve this by using a supervised learning algorithm able to learn with few images to facilitate the user task. We use a probabilistic model to recognise the objects and further we classify the scene based on their object occurrences. Experimental results are shown and evaluated to prove the validity of our proposal. Object recognition performance is compared to the approaches of He et al. (2004) and Marti et al. (2001) using their own datasets. Furthermore an unsupervised method is implemented in order to evaluate the advantages and disadvantages of our supervised classification approach versus an unsupervised one

Fast training of self organizing maps for the visual exploration of molecular compounds

Visual exploration of scientific data in life science area is a growing research field due to the large amount of available data. The Kohonen’s Self Organizing Map (SOM) is a widely used tool for visualization of multidimensional data. In this paper we present a fast learning algorithm for SOMs that uses a simulated annealing method to adapt the learning parameters. The algorithm has been adopted in a data analysis framework for the generation of similarity maps. Such maps provide an effective tool for the visual exploration of large and multi-dimensional input spaces. The approach has been applied to data generated during the High Throughput Screening of molecular compounds; the generated maps allow a visual exploration of molecules with similar topological properties. The experimental analysis on real world data from the National Cancer Institute shows the speed up of the proposed SOM training process in comparison to a traditional approach. The resulting visual landscape groups molecules with similar chemical properties in densely connected regions.

Application of personal computers to eddy current inspections

This presentation describes a system for measuring claddings as an example of the many possible advantages to be obtained by applying a personal computer to eddy current testing. A theoretical model and a learning algorithm are integrated into an instrument. They are supported in the PC, and serve to simplify and enhance multiparameter testing. The PC gives additional assistance by simplifying set-up procedures and data logging etc.

A camera based digit location and recognition system for garment tracking

Garment information tracking is required for clean room garment management. In this paper, we present a camera-based robust system with implementation of Optical Character Reconition (OCR) techniques to fulfill garment label recognition. In the system, a camera is used for image capturing; an adaptive thresholding algorithm is employed to generate binary images; Connected Component Labelling (CCL) is then adopted for object detection in the binary image as a part of finding the ROI (Region of Interest); Artificial Neural Networks (ANNs) with the BP (Back Propagation) learning algorithm are used for digit recognition; and finally the system is verified by a system database. The system has been tested. The results show that it is capable of coping with variance of lighting, digit twisting, background complexity, and font orientations. The system performance with association to the digit recognition rate has met the design requirement. It has achieved real-time and error-free garment information tracking during the testing.

Sparse model identification using orthogonal forward regression with basis pursuit and D-optimality

An efficient model identification algorithm for a large class of linear-in-the-parameters models is introduced that simultaneously optimises the model approximation ability, sparsity and robustness. The derived model parameters in each forward regression step are initially estimated via the orthogonal least squares (OLS), followed by being tuned with a new gradient-descent learning algorithm based on the basis pursuit that minimises the l(1) norm of the parameter estimate vector. The model subset selection cost function includes a D-optimality design criterion that maximises the determinant of the design matrix of the subset to ensure model robustness and to enable the model selection procedure to automatically terminate at a sparse model. The proposed approach is based on the forward OLS algorithm using the modified Gram-Schmidt procedure. Both the parameter tuning procedure, based on basis pursuit, and the model selection criterion, based on the D-optimality that is effective in ensuring model robustness, are integrated with the forward regression. As a consequence the inherent computational efficiency associated with the conventional forward OLS approach is maintained in the proposed algorithm. Examples demonstrate the effectiveness of the new approach.

Nonlinear model structure design and construction using orthogonal least squares and D-optimality design

A very efficient learning algorithm for model subset selection is introduced based on a new composite cost function that simultaneously optimizes the model approximation ability and model robustness and adequacy. The derived model parameters are estimated via forward orthogonal least squares, but the model subset selection cost function includes a D-optimality design criterion that maximizes the determinant of the design matrix of the subset to ensure the model robustness, adequacy, and parsimony of the final model. The proposed approach is based on the forward orthogonal least square (OLS) algorithm, such that new D-optimality-based cost function is constructed based on the orthogonalization process to gain computational advantages and hence to maintain the inherent advantage of computational efficiency associated with the conventional forward OLS approach. Illustrative examples are included to demonstrate the effectiveness of the new approach.

Nonlinear model structure detection using optimum experimental design and orthogonal least squares

A very efficient learning algorithm for model subset selection is introduced based on a new composite cost function that simultaneously optimizes the model approximation ability and model adequacy. The derived model parameters are estimated via forward orthogonal least squares, but the subset selection cost function includes an A-optimality design criterion to minimize the variance of the parameter estimates that ensures the adequacy and parsimony of the final model. An illustrative example is included to demonstrate the effectiveness of the new approach.