Automatic quantification of lobular inflammation and hepatocyte ballooning in nonalcoholic fatty liver disease liver biopsies.

Automatic quantification of cardinal histologic features of nonalcoholic fatty liver disease (NAFLD) may reduce human variability and allow continuous rather than semiquantitative assessment of injury. We recently developed an automated classifier that can detect and quantify macrosteatosis with greater than or equal to 95% precision and recall (sensitivity). Here, we report our early results on the classifier's performance in detecting lobular inflammation and hepatocellular ballooning. Automatic quantification of lobular inflammation and ballooning was performed on digital images of hematoxylin and eosin-stained slides of liver biopsy samples from 59 individuals with normal liver histology and varying severity of NAFLD. Two expert hepatopathologists scored liver biopsies according the nonalcoholic steatohepatitis clinical research network scoring system and provided annotations of lobular inflammation and hepatocyte ballooning on the digital images. The classifier had precision and recall of 70% and 49% for lobular inflammation, and 91% and 54% for hepatocyte ballooning. In addition, the classifier had an area under the curve of 95% for lobular inflammation and 98% for hepatocyte ballooning. The Spearman rank correlation coefficient for comparison with pathologist grades was 45.2% for lobular inflammation and 46% for hepatocyte ballooning. Our novel observations demonstrate that automatic quantification of cardinal NAFLD histologic lesions is feasible and offer promise for further development of automatic quantification as a potential aid to pathologists evaluating NAFLD biopsies in clinical practice and clinical trials.

Automatic classification of white regions in liver biopsies by supervised machine learning.

Automated assessment of histological features of non-alcoholic fatty liver disease (NAFLD) may reduce human variability and provide continuous rather than semiquantitative measurement of these features. As part of a larger effort, we perform automatic classification of steatosis, the cardinal feature of NAFLD, and other regions that manifest as white in images of hematoxylin and eosin-stained liver biopsy sections. These regions include macrosteatosis, central veins, portal veins, portal arteries, sinusoids and bile ducts. Digital images of hematoxylin and eosin-stained slides of 47 liver biopsies from patients with normal liver histology (n = 20) and NAFLD (n = 27) were obtained at 20× magnification. The images were analyzed using supervised machine learning classifiers created from annotations provided by two expert pathologists. The classification algorithm performs with 89% overall accuracy. It identified macrosteatosis, bile ducts, portal veins and sinusoids with high precision and recall (≥ 82%). Identification of central veins and portal arteries was less robust but still good. The accuracy of the classifier in identifying macrosteatosis is the best reported. The accurate automated identification of macrosteatosis achieved with this algorithm has useful clinical and research-related applications. The accurate detection of liver microscopic anatomical landmarks may facilitate important subsequent tasks, such as localization of other histological lesions according to liver microscopic anatomy.

Beyond captions: linking figures with abstract sentences in biomedical articles.

Although figures in scientific articles have high information content and concisely communicate many key research findings, they are currently under utilized by literature search and retrieval systems. Many systems ignore figures, and those that do not typically only consider caption text. This study describes and evaluates a fully automated approach for associating figures in the body of a biomedical article with sentences in its abstract. We use supervised methods to learn probabilistic language models, hidden Markov models, and conditional random fields for predicting associations between abstract sentences and figures. Three kinds of evidence are used: text in abstract sentences and figures, relative positions of sentences and figures, and the patterns of sentence/figure associations across an article. Each information source is shown to have predictive value, and models that use all kinds of evidence are more accurate than models that do not. Our most accurate method has an F1-score of 69% on a cross-validation experiment, is competitive with the accuracy of human experts, has significantly better predictive accuracy than state-of-the-art methods and enables users to access figures associated with an abstract sentence with an average of 1.82 fewer mouse clicks. A user evaluation shows that human users find our system beneficial. The system is available at http://FigureItOut.askHERMES.org.

Evaluation of the antigenic diversity of placenta-binding Plasmodium falciparum variants and the antibody repertoire among pregnant women.

Pregnant women are infected by specific variants of Plasmodium falciparum that adhere and accumulate in the placenta. Using serological and molecular approaches, we assessed the global antigenic diversity of surface antigens expressed by placenta-binding isolates to better understand immunity to malaria in pregnancy and evolution of polymorphisms and to inform vaccine development. We found that placenta-binding isolates originating from all major regions where malaria occurs were commonly recognized by antibodies in different populations of pregnant women. There was substantial antigenic overlap and sharing of epitopes between isolates, including isolates from distant geographic locations, suggesting that there are limitations to antigenic diversity; however, differences between populations and isolates were also seen. Many women had cross-reactive antibodies and/or a broad repertoire of antibodies to different isolates. Studying VAR2CSA as the major antigen expressed by placenta-binding isolates, we identified antibody epitopes encoded by variable sequence blocks in the DBL3 domain. Analysis of global var2csa DBL3 sequences demonstrated that there was extensive sharing of variable blocks between Africa, Asia, Papua New Guinea, and Latin America, which likely contributes to the high level of antigenic overlap between different isolates. However, there was also evidence of geographic clustering of sequences and differences in VAR2CSA sequences between populations. The results indicate that there is limited antigenic diversity in placenta-binding isolates and may explain why immunity to malaria in pregnancy can be achieved after exposure during one pregnancy. Inclusion of a limited number of variants in a candidate vaccine may be sufficient for broad population coverage, but geographic considerations may also have to be included in vaccine design.

Structural polymorphism and diversifying selection on the pregnancy malaria vaccine candidate VAR2CSA.

VAR2CSA is the main candidate for a pregnancy malaria vaccine, but vaccine development may be complicated by sequence polymorphism. Here, we obtained partial or full-length var2CSA sequences from 106 parasites and applied novel computational methods and three-dimensional modeling to investigate VAR2CSA geographic variation and selection pressure. Our analysis reveals structural patterns of VAR2CSA sequence variation in which polymorphic sites group into segments of limited diversity. Within these segments, two or three basic types characterize a substantial majority of the parasite samples. Comparison to the primate malaria Plasmodium reichenowi shows that these basic types have ancient origins. Globally, var2CSA genes are comprised of a mosaic of these ancestral polymorphic segments that have recombined extensively between var2CSA alleles. Three-dimensional modeling reveals that polymorphic segments concentrate in flexible loops at characteristic locations in the six VAR2CSA Duffy binding-like (DBL) adhesion domains. Individual DBL domain surfaces have distinct patterns of diversifying selection, suggesting that limited and differing portions of each DBL domain are targeted by host antibody. Since standard phylogenetic tree analysis is inadequate for highly recombining genes like var2CSA, we developed a novel phylogenetic approach that incorporates recombination and tracks new mutations in segment types. In the resulting tree, P. reichenowi is confirmed as an outlier and African and Asian P. falciparum isolates have slightly diverged. These findings validate a new approach to modeling protein evolution in the presence of frequent recombination and provide a clearer understanding of how var gene products function as immunoevasive binding ligands.

Connecting quantitative regulatory-network models to the genome.

MOTIVATION: An important task in computational biology is to infer, using background knowledge and high-throughput data sources, models of cellular processes such as gene regulation. Nachman et al. have developed an approach to inferring gene-regulatory networks that represents quantitative transcription rates, and simultaneously estimates both the kinetic parameters that govern these rates and the activity levels of unobserved regulators that control them. This approach is appealing in that it provides a more detailed and realistic description of how a gene's regulators influence its level of expression than alternative methods. We have developed an extension to this approach that involves representing and learning the key kinetic parameters as functions of features in the genomic sequence. The primary motivation for our approach is that it provides a more mechanistic representation of the regulatory relationships being modeled. RESULTS: We evaluate our approach using two Escherichia coli gene-expression data sets, with a particular focus on modeling the networks that are involved in controlling how E.coli regulates its response to the carbon source(s) available to it. Our results indicate that our sequence-based models provide predictive accuracy that is better than similar models without sequence-based parameters, and substantially better than a simple baseline. Moreover, our approach results in models that offer more explanatory power and biological insight than models without sequence-based parameters.

Predicting bacterial transcription units using sequence and expression data.

MOTIVATION: A key aspect of elucidating gene regulation in bacterial genomes is identifying the basic units of transcription. We present a method, based on probabilistic language models, that we apply to predict operons, promoters and terminators in the genome of Escherichia coli K-12. Our approach has two key properties: (i) it provides a coherent set of predictions for related regulatory elements of various types and (ii) it takes advantage of both DNA sequence and gene expression data, including expression measurements from inter-genic probes. RESULTS: Our experimental results show that we are able to predict operons and localize promoters and terminators with high accuracy. Moreover, our models that use both sequence and expression data are more accurate than those that use only one of these two data sources.

A Bayesian network approach to operon prediction.

MOTIVATION: In order to understand transcription regulation in a given prokaryotic genome, it is critical to identify operons, the fundamental units of transcription, in such species. While there are a growing number of organisms whose sequence and gene coordinates are known, by and large their operons are not known. RESULTS: We present a probabilistic approach to predicting operons using Bayesian networks. Our approach exploits diverse evidence sources such as sequence and expression data. We evaluate our approach on the Escherichia coli K-12 genome where our results indicate we are able to identify over 78% of its operons at a 10% false positive rate. Also, empirical evaluation using a reduced set of data sources suggests that our approach may have significant value for organisms that do not have as rich of evidence sources as E.coli. AVAILABILITY: Our E.coli K-12 operon predictions are available at http://www.biostat.wisc.edu/gene-regulation.