Pietro Lovato

Expression microarray classification using topic models

Classification of samples in expression microarray experiments represents a crucial task in bioinformatics and biomedicine. In this paper this scenario is addressed by employing a particular class of statistical approaches, called Topic Models. These models, firstly introduced in the text mining community, permit to extract from a set of objects (typically documents) an interpretable and rich description, based on an intermediate representation called topics (or processes). In this paper the expression microarray classification task is cast into this probabilistic context, providing a parallelism with the text mining domain and an interpretation. Two different topic models are investigated, namely the Probabilistic Latent Semantic Analysis (PLSA) and the Latent Dirichlet Allocation (LDA). An experimental evaluation of the proposed methodologies on three standard datasets confirms their effectiveness, also in comparison with other classification methodologies.

Biclustering of Expression Microarray Data with Topic Models

This paper presents an approach to extract biclusters from expression micro array data using topic models – a class of probabilistic models which allow to detect interpretable groups of highly correlated genes and samples. Starting from a topic model learned from the expression matrix, some automatic rules to extract biclusters are presented, which overcome the drawbacks of previous approaches. The methodology has been positively tested with synthetic benchmarks, as well as with a real experiment involving two different species of grape plants (Vitis vinifera and Vitis riparia).

Investigating Topic Models' Capabilities in Expression Microarray Data Classification

In recent years a particular class of probabilistic graphical models-called topic models-has proven to represent an useful and interpretable tool for understanding and mining microarray data. In this context, such models have been almost only applied in the clustering scenario, whereas the classification task has been disregarded by researchers. In this paper, we thoroughly investigate the use of topic models for classification of microarray data, starting from ideas proposed in other fields (e.g., computer vision). A classification scheme is proposed, based on highly interpretable features extracted from topic models, resulting in a hybrid generative-discriminative approach; an extensive experimental evaluation, involving 10 different literature benchmarks, confirms the suitability of the topic models for classifying expression microarray data.

Faved! Biometrics: Tell Me Which Image You Like and I'll Tell You Who You Are

This paper builds upon the belief that every human being has a built-in image aesthetic evaluation system. This sort of personal aesthetics mostly follows certain aesthetic rules widely studied in image aesthetics (e.g., rules of thirds, colorfulness, etc.), though it likely contains some innate, unique preferences. This paper is a proof of concept of this intuition, presenting personal aesthetics as a novel behavioral biometrical trait. In our scenario, personal aesthetics activate when an individual is presented with a set of photos he may like or dislike. The goal is to distill and encode the uniqueness of his visual preferences into a compact template. To this aim, we extract a pool of low- and high-level state-of-the-art image features from a set of Flickr images preferred by a user, feeding them successively into a LASSO regressor. LASSO highlights the most discriminant cues for the individual, allowing authentication and recognition tasks. The results are surprising given only 1 image as test. We can match the user identity against a gallery of 200 individuals definitely much better than chance. Using 20 images (all preferred by a single user) as a biometrical trait, we reach an AUC of 96%, considering the cumulative matching characteristic curve. Extensive experiments also support the interpretability of our approach, effectively modeling what is the “what we like” that distinguishes us from others.

Feature selection using counting grids: application to microarray data

In this paper a novel feature selection scheme is proposed, which exploits the potentialities of a recent probabilistic generative model, the Counting Grid. This model is able to cluster together similar observations, highlighting the compactness of a class and its underlying structure. The proposed feature selection scheme is applied to the expression microarray scenario, a peculiar context with very few patterns and a huge number of features. Experiments on benchmark datasets show that the proposed approach is effective and stable, assessing state-of-the-art classification accuracies.

A bioinformatics approach to 2D shape classification

An alternative interaction between Pattern Recognition and Bioinformatics is studied.2D shape classification is faced using biological sequence analysis approaches.Classification results are competitive with literature.Other bioinformatics tools are used for understanding and interpretation. In the past, the huge and profitable interaction between Pattern Recognition and biology/bioinformatics was mainly unidirectional, namely targeted at applying PR tools and ideas to analyse biological data. In this paper we investigate an alternative approach, which exploits bioinformatics solutions to solve PR problems: in particular, we address the 2D shape classification problem using classical biological sequence analysis approaches - for which a vast amount of tools and solutions have been developed and improved in more than 40 years of research. First, we highlight the similarities between 2D shapes and biological sequences, then we propose three methods to encode a shape as a biological sequence. Given the encoding, we can employ standard biological sequence analysis tools to derive a similarity, which can be exploited in a nearest neighbor framework. Classification results, obtained on 5 standard datasets, confirm the potentials of the proposed unconventional interaction between PR and bioinformatics. Moreover, we provide some evidences of how it is possible to exploit other bioinformatics concepts and tools to interpret data and results, confirming the flexibility of the proposed framework.

S-BLOSUM: Classification of 2D Shapes with Biological Sequence Alignment

Recent works investigated the possibility to design solutions for pattern recognition problems by exploiting the huge amount of work done in bioinformatics. If the pattern recognition problem is cast in biological terms, then a huge range of algorithms, exploitable for classification, detection, visualization, etc. can be effectively borrowed. In this paper, we exploit biological sequence alignment tools to classify 2D shapes, tailoring the biological parameters of these tools to account for the different semantic of the 2D shape scenario. In particular, we propose a novel substitution matrix, which is the crucial parameter determining the sequence alignment solution. The new matrix, called S-BLOSUM, learns the rates of matches/mismatches in conserved portions of shapes belonging to the same category, and incorporates prior knowledge on the chosen representation for the 2D shape. On one hand, the experimental evaluation showed that the S-BLOSUM provides a significant improvement over the biological counterpart (BLOSUM), on the other hand, classification results prove that our approach is competitive with respect to the state of the art.

2D shapes classification using BLAST

This paper presents a novel 2D shape classification approach, which exploits in this context the huge amount of work carried out by bioinformaticians in the biological sequence analysis research field. In particular, in the approach presented here, we propose to encode shapes as biological sequences, employing the widely known sequence alignment tool called BLAST (Basic Local Alignment Search Tool) to devise a similarity score, used in a nearest neighbour scenario. Obtained results on standard datasets show the feasibility of the proposed approach.

A comparison on score spaces for expression microarray data classification

In this paper an empirical evaluation of different generative scores for expression microarray data classification is proposed. Score spaces represent a quite recent trend in the machine learning community, taking the best of both generative and discriminative classification paradigms. The scores are extracted from topic models, a class of highly interpretable probabilistic tools whose utility in the microarray classification context has been recently assessed. The experimental evaluation, performed on 3 literature datasets and with 7 score spaces, demonstrates the viability of the proposed scheme and, for the first time, it compares pros and cons of each space.

Supranormal thymic output up to 2 decades after HIV-1 infection.

Objectives:AIDS is caused by CD4+ T-cell depletion. Although combination antiretroviral therapy can restore blood T-cell numbers, the clonal diversity of the reconstituting cells, critical for immunocompetence, is not well defined. Methods:We performed an extensive analysis of parameters of thymic function in perinatally HIV-1-infected (n = 39) and control (n = 28) participants ranging from 13 to 23 years of age. CD4+ T cells including naive (CD27+ CD45RA+) and recent thymic emigrant (RTE) (CD31+/CD45RA+) cells, were quantified by flow cytometry. Deep sequencing was used to examine T-cell receptor (TCR) sequence diversity in sorted RTE CD4+ T cells. Results:Infected participants had reduced CD4+ T-cell levels with predominant depletion of the memory subset and preservation of naive cells. RTE CD4+ T-cell levels were normal in most infected individuals, and enhanced thymopoiesis was indicated by higher proportions of CD4+ T cells containing TCR recombination excision circles. Memory CD4+ T-cell depletion was highly associated with CD8+ T-cell activation in HIV-1-infected persons and plasma interlekin-7 levels were correlated with naive CD4+ T cells, suggesting activation-driven loss and compensatory enhancement of thymopoiesis. Deep sequencing of CD4+ T-cell receptor sequences in well compensated infected persons demonstrated supranormal diversity, providing additional evidence of enhanced thymic output. Conclusion:Despite up to two decades of infection, many individuals have remarkable thymic reserve to compensate for ongoing CD4+ T-cell loss, although there is ongoing viral replication and immune activation despite combination antiretroviral therapy. The longer term sustainability of this physiology remains to be determined.