Tomer Hertz

Increased HIV-1 vaccine efficacy against viruses with genetic signatures in Env V2

The RV144 trial demonstrated 31% vaccine efficacy at preventing human immunodeficiency virus (HIV)-1 infection. Antibodies against the HIV-1 envelope variable loops 1 and 2 (Env V1 and V2) correlated inversely with infection risk. We proposed that vaccine-induced immune responses against V1/V2 would have a selective effect against, or sieve, HIV-1 breakthrough viruses. A total of 936 HIV-1 genome sequences from 44 vaccine and 66 placebo recipients were examined. We show that vaccine-induced immune responses were associated with two signatures in V2 at amino acid positions 169 and 181. Vaccine efficacy against viruses matching the vaccine at position 169 was 48% (confidence interval 18% to 66%; P = 0.0036), whereas vaccine efficacy against viruses mismatching the vaccine at position 181 was 78% (confidence interval 35% to 93%; P = 0.0028). Residue 169 is in a cationic glycosylated region recognized by broadly neutralizing and RV144-derived antibodies. The predicted distance between the two signature sites (21 ± 7 Å) and their match/mismatch dichotomy indicate that multiple factors may be involved in the protection observed in RV144. Genetic signatures of RV144 vaccination in V2 complement the finding of an association between high V1/V2-binding antibodies and reduced risk of HIV-1 acquisition, and provide evidence that vaccine-induced V2 responses plausibly had a role in the partial protection conferred by the RV144 regimen.

Adjustment Learning and Relevant Component Analysis

We propose a new learning approach for image retrieval, which we call adjustment learning, and demonstrate its use for face recognition and color matching. Our approach is motivated by a frequently encountered problem, namely, that variability in the original data representation which is not relevant to the task may interfere with retrieval and make it very difficult. Our key observation is that in real applications of image retrieval, data sometimes comes in small chunks - small subsets of images that come from the same (but unknown) class. This is the case, for example, when a query is presented via a short video clip. We call these groups chunklets, and we call the paradigm which uses chunklets for unsupervised learning adjustment learning. Within this paradigm we propose a linear scheme, which we call Relevant Component Analysis; this scheme uses the information in such chunklets to reduce irrelevant variability in the data while amplifying relevant variability. We provide results using our method on two problems: face recognition (using a database publicly available on the web), and visual surveillance (using our own data). In the latter application chunklets are obtained automatically from the data without the need of supervision.

Learning distance functions for image retrieval

Image retrieval critically relies on the distance function used to compare a query image to images in the database. We suggest learning such distance functions by training binary classifiers with margins, where the classifiers are defined over the product space of pairs of images. The classifiers are trained to distinguish between pairs in which the images are from the same class and pairs, which contain images from different classes. The signed margin is used as a distance function. We explore several variants of this idea, based on using SVM and boosting algorithms as product space classifiers. Our main contribution is a distance learning method, which combines boosting hypotheses over the product space with a weak learner based on partitioning the original feature space. The weak learner used is a Gaussian mixture model computed using a constrained EM algorithm, where the constraints are equivalence constraints on pairs of data points. This approach allows us to incorporate unlabeled data into the training process. Using some benchmark databases from the UCI repository, we show that our margin based methods significantly outperform existing metric learning methods, which are based an learning a Mahalanobis distance. We then show comparative results of image retrieval in a distributed learning paradigm, using two databases: a large database of facial images (YaleB), and a database of natural images taken from a commercial CD. In both cases our GMM based boosting method outperforms all other methods, and its generalization to unseen classes is superior.

Boosting margin based distance functions for clustering

The performance of graph based clustering methods critically depends on the quality of the distance function used to compute similarities between pairs of neighboring nodes. In this paper we learn distance functions by training binary classifiers with margins. The classifiers are defined over the product space of pairs of points and are trained to distinguish whether two points come from the same class or not. The signed margin is used as the distance value. Our main contribution is a distance learning method (DistBoost), which combines boosting hypotheses over the product space with a weak learner based on partitioning the original feature space. Each weak hypothesis is a Gaussian mixture model computed using a semi-supervised constrained EM algorithm, which is trained using both unlabeled and labeled data. We also consider SVM and decision trees boosting as margin based classifiers in the product space. We experimentally compare the margin based distance functions with other existing metric learning methods, and with existing techniques for the direct incorporation of constraints into various clustering algorithms. Clustering performance is measured on some benchmark databases from the UCI repository, a sample from the MNIST database, and a data set of color images of animals. In most cases the DistBoost algorithm significantly and robustly outperformed its competitors.

Mucosal Immune Responses Predict Clinical Outcomes during Influenza Infection Independently of Age and Viral Load

RATIONALE Children are an at-risk population for developing complications following influenza infection, but immunologic correlates of disease severity are not understood. We hypothesized that innate cellular immune responses at the site of infection would correlate with disease outcome. OBJECTIVES To test the immunologic basis of severe illness during natural influenza virus infection of children and adults at the site of infection. METHODS An observational cohort study with longitudinal sampling of peripheral and mucosal sites in 84 naturally influenza-infected individuals, including infants. Cellular responses, viral loads, and cytokines were quantified from nasal lavages and blood, and correlated to clinical severity. MEASUREMENTS AND MAIN RESULTS We show for the first time that although viral loads in children and adults were similar, innate responses in the airways were stronger in children and varied considerably between plasma and site of infection. Adjusting for age and viral load, an innate immune profile characterized by increased nasal lavage monocyte chemotactic protein-3, IFN-α2, and plasma IL-10 levels at enrollment predicted progression to severe disease. Increased plasma IL-10, monocyte chemotactic protein-3, and IL-6 levels predicted hospitalization. This inflammatory cytokine production correlated significantly with monocyte localization from the blood to the site of infection, with conventional monocytes positively correlating with inflammation. Increased frequencies of CD14(lo) monocytes were in the airways of participants with lower inflammatory cytokine levels. CONCLUSIONS An innate profile was identified that correlated with disease progression independent of viral dynamics and age. The airways and blood displayed dramatically different immune profiles emphasizing the importance of cellular migration and localized immune phenotypes.

Quantifiable predictive features define epitope-specific T cell receptor repertoires

T cells are defined by a heterodimeric surface receptor, the T cell receptor (TCR), that mediates recognition of pathogen-associated epitopes through interactions with peptide and major histocompatibility complexes (pMHCs). TCRs are generated by genomic rearrangement of the germline TCR locus, a process termed V(D)J recombination, that has the potential to generate marked diversity of TCRs (estimated to range from 1015 (ref. 1) to as high as 1061 (ref. 2) possible receptors). Despite this potential diversity, TCRs from T cells that recognize the same pMHC epitope often share conserved sequence features, suggesting that it may be possible to predictively model epitope specificity. Here we report the in-depth characterization of ten epitope-specific TCR repertoires of CD8+ T cells from mice and humans, representing over 4,600 in-frame single-cell-derived TCRαβ sequence pairs from 110 subjects. We developed analytical tools to characterize these epitope-specific repertoires: a distance measure on the space of TCRs that permits clustering and visualization, a robust repertoire diversity metric that accommodates the low number of paired public receptors observed when compared to single-chain analyses, and a distance-based classifier that can assign previously unobserved TCRs to characterized repertoires with robust sensitivity and specificity. Our analyses demonstrate that each epitope-specific repertoire contains a clustered group of receptors that share core sequence similarities, together with a dispersed set of diverse ‘outlier’ sequences. By identifying shared motifs in core sequences, we were able to highlight key conserved residues driving essential elements of TCR recognition. These analyses provide insights into the generalizable, underlying features of epitope-specific repertoires and adaptive immune recognition.

PepDist: a new framework for protein-peptide binding prediction based on learning peptide distance functions.

BackgroundMany different aspects of cellular signalling, trafficking and targeting mechanisms are mediated by interactions between proteins and peptides. Representative examples are MHC-peptide complexes in the immune system. Developing computational methods for protein-peptide binding prediction is therefore an important task with applications to vaccine and drug design.MethodsPrevious learning approaches address the binding prediction problem using traditional margin based binary classifiers. In this paper we propose PepDist: a novel approach for predicting binding affinity. Our approach is based on learning peptide-peptide distance functions. Moreover, we suggest to learn a single peptide-peptide distance function over an entire family of proteins (e.g. MHC class I). This distance function can be used to compute the affinity of a novel peptide to any of the proteins in the given family. In order to learn these peptide-peptide distance functions, we formalize the problem as a semi-supervised learning problem with partial information in the form of equivalence constraints. Specifically, we propose to use DistBoost [1, 2], which is a semi-supervised distance learning algorithm.ResultsWe compare our method to various state-of-the-art binding prediction algorithms on MHC class I and MHC class II datasets. In almost all cases, our method outperforms all of its competitors. One of the major advantages of our novel approach is that it can also learn an affinity function over proteins for which only small amounts of labeled peptides exist. In these cases, our methods performance gain, when compared to other computational methods, is even more pronounced. We have recently uploaded the PepDist webserver which provides binding prediction of peptides to 35 different MHC class I alleles. The webserver which can be found at http://www.pepdist.cs.huji.ac.il is powered by a prediction engine which was trained using the framework presented in this paper.ConclusionThe results obtained suggest that learning a single distance function over an entire family of proteins achieves higher prediction accuracy than learning a set of binary classifiers for each of the proteins separately. We also show the importance of obtaining information on experimentally determined non-binders. Learning with real non-binders generalizes better than learning with randomly generated peptides that are assumed to be non-binders. This suggests that information about non-binding peptides should also be published and made publicly available.

Machine learning competition in immunology – Prediction of HLA class I binding peptides

Experimental studies of immune system and related applications such as characterization of immune responses against pathogens, vaccine design, or optimization of therapies are combinatorially complex, time-consuming and expensive. The main methods for large-scale identification of T-cell epitopes from pathogens or cancer proteomes involve either reverse immunology or high-throughput mass spectrometry (HTMS). Reverse immunology approaches involve pre-screening of proteomes by computational algorithms, followed by experimental validation of selected targets (Mora et al., 2006; De Groot et al., 2008; Larsen et al., 2010). HTMS involves HLA typing, immunoaffinity chromatography of HLA molecules, HLA extraction, and chromatography combined with tandem mass spectrometry, followed by the application of computational algorithms for peptide characterization (Bassani-Sternberg et al., 2010). Hundreds of naturally processed HLA class I associated peptides have been identified in individual studies using HTMS in normal (Escobar et al., 2008), cancer (Antwi et al., 2009; Bassani-Sternberg et al., 2010), autoimmunity-related (Ben Dror et al., 2010), and infected samples (Wahl et al, 2010). Computational algorithms are essential steps in highthroughput identification of T-cell epitope candidates using both reverse immunology and HTMS approaches. Peptide binding to MHC molecules is the single most selective step in defining T cell epitope and the accuracy of computational algorithms for prediction of peptide binding, therefore, determines the accuracy of the overall method. Computational predictions of peptide binding to HLA, both class I and class II, use a variety of algorithms ranging from binding motifs to advanced machine learning techniques (Brusic et al., 2004; Lafuente and Reche, 2009) and standards for their

Mapping the Landscape of Host-Pathogen Coevolution: HLA Class I Binding and Its Relationship with Evolutionary Conservation in Human and Viral Proteins

ABSTRACT The high diversity of HLA binding preferences has been driven by the sequence diversity of short segments of relevant pathogenic proteins presented by HLA molecules to the immune system. To identify possible commonalities in HLA binding preferences, we quantify these using a novel measure termed “targeting efficiency,” which captures the correlation between HLA-peptide binding affinities and the conservation of the targeted proteomic regions. Analysis of targeting efficiencies for 95 HLA class I alleles over thousands of human proteins and 52 human viruses indicates that HLA molecules preferentially target conserved regions in these proteomes, although the arboviral Flaviviridae are a notable exception where nonconserved regions are preferentially targeted by most alleles. HLA-A alleles and several HLA-B alleles that have maintained close sequence identity with chimpanzee homologues target conserved human proteins and DNA viruses such as Herpesviridae and Adenoviridae most efficiently, while all HLA-B alleles studied efficiently target RNA viruses. These patterns of host and pathogen specialization are both consistent with coevolutionary selection and functionally relevant in specific cases; for example, preferential HLA targeting of conserved proteomic regions is associated with improved outcomes in HIV infection and with protection against dengue hemorrhagic fever. Efficiency analysis provides a novel perspective on the coevolutionary relationship between HLA class I molecular diversity, self-derived peptides that shape T-cell immunity through ontogeny, and the broad range of viruses that subsequently engage with the adaptive immune response.

Enhancing image and video retrieval: learning via equivalence constraints

The paper is about learning using partial information in the form of equivalence constraints. Equivalence constraints provide relational information about the labels of data points, rather than the labels themselves. Our work is motivated by the observation that in many real life applications partial information about the data can be obtained with very little cost. For example, in video indexing we may want to use the fact that a sequence of faces obtained from successive frames in roughly the same location is likely to contain the same unknown individual. Learning using equivalence constraints is different from learning using labels and poses new technical challenges. In this paper we present three novel methods for clustering and classification, which use equivalence constraints. We provide results of our methods on a distributed image querying system that works on a large facial image database, and on the clustering and retrieval of surveillance data. Our results show that we can significantly improve the performance of image retrieval by taking advantage of such assumptions as temporal continuity in the data. Significant improvement is also obtained by making the users of the system take the role of distributed teachers, which reduces the need for expensive labeling by paid human labor.