Large-scale investigation of weakly-supervised deep learning for the fine-grained semantic indexing of biomedical literature.

OBJECTIVE: Semantic indexing of biomedical literature is usually done at the level of MeSH descriptors with several related but distinct biomedical concepts often grouped together and treated as a single topic. This study proposes a new method for the automated refinement of subject annotations at the level of MeSH concepts. METHODS: Lacking labelled data, we rely on weak supervision based on concept occurrence in the abstract of an article, which is also enhanced by dictionary-based heuristics. In addition, we investigate deep learning approaches, making design choices to tackle the particular challenges of this task. The new method is evaluated on a large-scale retrospective scenario, based on concepts that have been promoted to descriptors. RESULTS: In our experiments concept occurrence was the strongest heuristic achieving a macro-F1 score of about 0.63 across several labels. The proposed method improved it further by more than 4pp. CONCLUSION: The results suggest that concept occurrence is a strong heuristic for refining the coarse-grained labels at the level of MeSH concepts and the proposed method improves it further.

Multi-probe attention neural network for COVID-19 semantic indexing.

BACKGROUND: The COVID-19 pandemic has increasingly accelerated the publication pace of scientific literature. How to efficiently curate and index this large amount of biomedical literature under the current crisis is of great importance. Previous literature indexing is mainly performed by human experts using Medical Subject Headings (MeSH), which is labor-intensive and time-consuming. Therefore, to alleviate the expensive time consumption and monetary cost, there is an urgent need for automatic semantic indexing technologies for the emerging COVID-19 domain. RESULTS: In this research, to investigate the semantic indexing problem for COVID-19, we first construct the new COVID-19 Semantic Indexing dataset, which consists of more than 80 thousand biomedical articles. We then propose a novel semantic indexing framework based on the multi-probe attention neural network (MPANN) to address the COVID-19 semantic indexing problem. Specifically, we employ a k-nearest neighbour based MeSH masking approach to generate candidate topic terms for each input article. We encode and feed the selected candidate terms as well as other contextual information as probes into the downstream attention-based neural network. Each semantic probe carries specific aspects of biomedical knowledge and provides informatively discriminative features for the input article. After extracting the semantic features at both term-level and document-level through the attention-based neural network, MPANN adopts a linear multi-view classifier to conduct the final topic prediction for COVID-19 semantic indexing. CONCLUSION: The experimental results suggest that MPANN promises to represent the semantic features of biomedical texts and is effective in predicting semantic topics for COVID-19 related biomedical articles.

Evaluation of Sirtuin-3 probe quality and co-expressed genes using literature cohesion.

BACKGROUND: Gene co-expression studies can provide important insights into molecular and cellular signaling pathways. The GeneNetwork database is a unique resource for co-expression analysis using data from a variety of tissues across genetically distinct inbred mice. However, extraction of biologically meaningful co-expressed gene sets is challenging due to variability in microarray platforms, probe quality, normalization methods, and confounding biological factors. In this study, we tested whether literature derived functional cohesion could be used as an objective metric in lieu of 'ground truth' to evaluate the quality of probes and microarray datasets. RESULTS: We examined Sirtuin-3 (Sirt3) co-expressed gene sets extracted from either liver or brain tissues of BXD recombinant inbred mice in the GeneNetwork database. Depending on the microarray platform, there were as many as 26 probes that targeted different regions of Sirt3 primary transcript. Co-expressed gene sets (ranging from 100-1000 genes) associated with each Sirt3 probe were evaluated using the previously developed literature-derived cohesion p-value (LPv) and benchmarked against 'gold standards' derived from proteomic studies or Gene Ontology classifications. We found that the maximal F-measure was obtained at an average window size of 535 genes. Using set size of 500 genes, the Pearson correlations between LPv and F-measure as well as between LPv and mitochondrial gene enrichment p-values were 0.90 and 0.93, respectively. Importantly, we found that the LPv approach can distinguish high quality Sirt3 probes. Analysis of the most functionally cohesive Sirt3 co-expressed gene set revealed core metabolic pathways that were shared between hippocampus and liver as well as distinct pathways which were unique to each tissue. These results are consistent with other studies that suggest Sirt3 is a key metabolic regulator and has distinct functions in energy-producing vs. energy-demanding tissues. CONCLUSIONS: Our results provide proof-of-concept that literature cohesion analysis is useful for evaluating the quality of probes and microarray datasets, particularly when experimentally derived gold standards are unavailable. Our approach would enable researchers to rapidly identify biologically meaningful co-expressed gene sets and facilitate discovery from high throughput genomic data.

Biomedical semantic indexing by deep neural network with multi-task learning.

BACKGROUND: Biomedical semantic indexing is important for information retrieval and many other research fields in bioinformatics. It annotates biomedical citations with Medical Subject Headings. In face of unbalanced category distribution in the training data, sampling methods are difficult to apply for semantic indexing task. RESULTS: In this paper, we present a novel deep serial multi-task learning model. The primary task treats the biomedical semantic indexing as a multi-label text classification issue that considers the relations of the labels. The auxiliary task is a regression task that predicts the MeSH number of the citation and provides hints for the network to make it converge faster. The experimental results on the BioASQ-Task5A open dataset show that our model outperforms the state-of-the-art solution "MTI", proposed by the US National Library of Medicine. Further, it not only achieves the highest precision among all the solutions in BioASQ-Task5A but also has faster convergence speed compared with some naive deep learning methods. CONCLUSIONS: Rather than parallel in an ordinary multi-task structure, the tasks in our model are serial and tightly coupled. It can achieve satisfied performance without any handcrafted feature.

Embedding of semantic predications.

This paper concerns the generation of distributed vector representations of biomedical concepts from structured knowledge, in the form of subject-relation-object triplets known as semantic predications. Specifically, we evaluate the extent to which a representational approach we have developed for this purpose previously, known as Predication-based Semantic Indexing (PSI), might benefit from insights gleaned from neural-probabilistic language models, which have enjoyed a surge in popularity in recent years as a means to generate distributed vector representations of terms from free text. To do so, we develop a novel neural-probabilistic approach to encoding predications, called Embedding of Semantic Predications (ESP), by adapting aspects of the Skipgram with Negative Sampling (SGNS) algorithm to this purpose. We compare ESP and PSI across a number of tasks including recovery of encoded information, estimation of semantic similarity and relatedness, and identification of potentially therapeutic and harmful relationships using both analogical retrieval and supervised learning. We find advantages for ESP in some, but not all of these tasks, revealing the contexts in which the additional computational work of neural-probabilistic modeling is justified.

Prioritization, clustering and functional annotation of MicroRNAs using latent semantic indexing of MEDLINE abstracts.

BACKGROUND: The amount of scientific information about MicroRNAs (miRNAs) is growing exponentially, making it difficult for researchers to interpret experimental results. In this study, we present an automated text mining approach using Latent Semantic Indexing (LSI) for prioritization, clustering and functional annotation of miRNAs. RESULTS: For approximately 900 human miRNAs indexed in miRBase, text documents were created by concatenating titles and abstracts of MEDLINE citations which refer to the miRNAs. The documents were parsed and a weighted term-by-miRNA frequency matrix was created, which was subsequently factorized via singular value decomposition to extract pair-wise cosine values between the term (keyword) and miRNA vectors in reduced rank semantic space. LSI enables derivation of both explicit and implicit associations between entities based on word usage patterns. Using miR2Disease as a gold standard, we found that LSI identified keyword-to-miRNA relationships with high accuracy. In addition, we demonstrate that pair-wise associations between miRNAs can be used to group them into categories which are functionally aligned. Finally, term ranking by querying the LSI space with a group of miRNAs enabled annotation of the clusters with functionally related terms. CONCLUSIONS: LSI modeling of MEDLINE abstracts provides a robust and automated method for miRNA related knowledge discovery. The latest collection of miRNA abstracts and LSI model can be accessed through the web tool miRNA Literature Network (miRLiN) at http://bioinfo.memphis.edu/mirlin .

Identifying plausible adverse drug reactions using knowledge extracted from the literature.

Pharmacovigilance involves continually monitoring drug safety after drugs are put to market. To aid this process; algorithms for the identification of strongly correlated drug/adverse drug reaction (ADR) pairs from data sources such as adverse event reporting systems or Electronic Health Records have been developed. These methods are generally statistical in nature, and do not draw upon the large volumes of knowledge embedded in the biomedical literature. In this paper, we investigate the ability of scalable Literature Based Discovery (LBD) methods to identify side effects of pharmaceutical agents. The advantage of LBD methods is that they can provide evidence from the literature to support the plausibility of a drug/ADR association, thereby assisting human review to validate the signal, which is an essential component of pharmacovigilance. To do so, we draw upon vast repositories of knowledge that has been extracted from the biomedical literature by two Natural Language Processing tools, MetaMap and SemRep. We evaluate two LBD methods that scale comfortably to the volume of knowledge available in these repositories. Specifically, we evaluate Reflective Random Indexing (RRI), a model based on concept-level co-occurrence, and Predication-based Semantic Indexing (PSI), a model that encodes the nature of the relationship between concepts to support reasoning analogically about drug-effect relationships. An evaluation set was constructed from the Side Effect Resource 2 (SIDER2), which contains known drug/ADR relations, and models were evaluated for their ability to "rediscover" these relations. In this paper, we demonstrate that both RRI and PSI can recover known drug-adverse event associations. However, PSI performed better overall, and has the additional advantage of being able to recover the literature underlying the reasoning pathways it used to make its predictions.

The road from manual to automatic semantic indexing of biomedical literature: a 10 years journey.

Biomedical experts are facing challenges in keeping up with the vast amount of biomedical knowledge published daily. With millions of citations added to databases like MEDLINE/PubMed each year, efficiently accessing relevant information becomes crucial. Traditional term-based searches may lead to irrelevant or missed documents due to homonyms, synonyms, abbreviations, or term mismatch. To address this, semantic search approaches employing predefined concepts with associated synonyms and relations have been used to expand query terms and improve information retrieval. The National Library of Medicine (NLM) plays a significant role in this area, indexing citations in the MEDLINE database with topic descriptors from the Medical Subject Headings (MeSH) thesaurus, enabling advanced semantic search strategies to retrieve relevant citations, despite synonymy, and polysemy of biomedical terms. Over time, advancements in semantic indexing have been made, with Machine Learning facilitating the transition from manual to automatic semantic indexing in the biomedical literature. The paper highlights the journey of this transition, starting with manual semantic indexing and the initial efforts toward automatic indexing. The BioASQ challenge has served as a catalyst in revolutionizing the domain of semantic indexing, further pushing the boundaries of efficient knowledge retrieval in the biomedical field.

G Protein-Coupled Receptor Systems and Their Role in Cellular Senescence.

Aging is a complex biological process that is inevitable for nearly all organisms. Aging is the strongest risk factor for development of multiple neurodegenerative disorders, cancer and cardiovascular disorders. Age-related disease conditions are mainly caused by the progressive degradation of the integrity of communication systems within and between organs. This is in part mediated by, i) decreased efficiency of receptor signaling systems and ii) an increasing inability to cope with stress leading to apoptosis and cellular senescence. Cellular senescence is a natural process during embryonic development, more recently it has been shown to be also involved in the development of aging disorders and is now considered one of the major hallmarks of aging. G-protein-coupled receptors (GPCRs) comprise a superfamily of integral membrane receptors that are responsible for cell signaling events involved in nearly every physiological process. Recent advances in the molecular understanding of GPCR signaling complexity have expanded their therapeutic capacity tremendously. Emerging data now suggests the involvement of GPCRs and their associated proteins in the development of cellular senescence. With the proven efficacy of therapeutic GPCR targeting, it is reasonable to now consider GPCRs as potential platforms to control cellular senescence and the consequently, age-related disorders.

Search and Graph Database Technologies for Biomedical Semantic Indexing: Experimental Analysis.

BACKGROUND: Biomedical semantic indexing is a very useful support tool for human curators in their efforts for indexing and cataloging the biomedical literature. OBJECTIVE: The aim of this study was to describe a system to automatically assign Medical Subject Headings (MeSH) to biomedical articles from MEDLINE. METHODS: Our approach relies on the assumption that similar documents should be classified by similar MeSH terms. Although previous work has already exploited the document similarity by using a k-nearest neighbors algorithm, we represent documents as document vectors by search engine indexing and then compute the similarity between documents using cosine similarity. Once the most similar documents for a given input document are retrieved, we rank their MeSH terms to choose the most suitable set for the input document. To do this, we define a scoring function that takes into account the frequency of the term into the set of retrieved documents and the similarity between the input document and each retrieved document. In addition, we implement guidelines proposed by human curators to annotate MEDLINE articles; in particular, the heuristic that says if 3 MeSH terms are proposed to classify an article and they share the same ancestor, they should be replaced by this ancestor. The representation of the MeSH thesaurus as a graph database allows us to employ graph search algorithms to quickly and easily capture hierarchical relationships such as the lowest common ancestor between terms. RESULTS: Our experiments show promising results with an F1 of 69% on the test dataset. CONCLUSIONS: To the best of our knowledge, this is the first work that combines search and graph database technologies for the task of biomedical semantic indexing. Due to its horizontal scalability, ElasticSearch becomes a real solution to index large collections of documents (such as the bibliographic database MEDLINE). Moreover, the use of graph search algorithms for accessing MeSH information could provide a support tool for cataloging MEDLINE abstracts in real time.

Large-scale online semantic indexing of biomedical articles via an ensemble of multi-label classification models.

BACKGROUND: In this paper we present the approach that we employed to deal with large scale multi-label semantic indexing of biomedical papers. This work was mainly implemented within the context of the BioASQ challenge (2013-2017), a challenge concerned with biomedical semantic indexing and question answering. METHODS: Our main contribution is a MUlti-Label Ensemble method (MULE) that incorporates a McNemar statistical significance test in order to validate the combination of the constituent machine learning algorithms. Some secondary contributions include a study on the temporal aspects of the BioASQ corpus (observations apply also to the BioASQ's super-set, the PubMed articles collection) and the proper parametrization of the algorithms used to deal with this challenging classification task. RESULTS: The ensemble method that we developed is compared to other approaches in experimental scenarios with subsets of the BioASQ corpus giving positive results. In our participation in the BioASQ challenge we obtained the first place in 2013 and the second place in the four following years, steadily outperforming MTI, the indexing system of the National Library of Medicine (NLM). CONCLUSIONS: The results of our experimental comparisons, suggest that employing a statistical significance test to validate the ensemble method's choices, is the optimal approach for ensembling multi-label classifiers, especially in contexts with many rare labels.

Asymmetric latent semantic indexing for gene expression experiments visualization.

We propose a new method to visualize gene expression experiments inspired by the latent semantic indexing technique originally proposed in the textual analysis context. By using the correspondence word-gene document-experiment, we define an asymmetric similarity measure of association for genes that accounts for potential hierarchies in the data, the key to obtain meaningful gene mappings. We use the polar decomposition to obtain the sources of asymmetry of the similarity matrix, which are later combined with previous knowledge. Genetic classes of genes are identified by means of a mixture model applied in the genes latent space. We describe the steps of the procedure and we show its utility in the Human Cancer dataset.

Semantic mapping reveals distinct patterns in descriptions of social relations in adults with autism spectrum disorder.

Adults with autism spectrum disorder (ASD) may describe other individuals differently compared with typical adults. In this study, we first asked participants to describe closely related individuals such as parents and close friends with 10 positive and 10 negative characteristics. We then used standard natural language processing methods to digitize and visualize these descriptions. The complex patterns of these descriptive sentences exhibited a difference in semantic space between individuals with ASD and control participants. Machine learning algorithms were able to automatically detect and discriminate between these two groups. Furthermore, we showed that these descriptive sentences from adults with ASD exhibited fewer connections as defined by word-word co-occurrences in descriptions, and these connections in words formed a less "small-world" like network. Autism Res 2016, 9: 846-853. © 2015 International Society for Autism Research, Wiley Periodicals, Inc.

Large scale biomedical texts classification: a kNN and an ESA-based approaches.

BACKGROUND: With the large and increasing volume of textual data, automated methods for identifying significant topics to classify textual documents have received a growing interest. While many efforts have been made in this direction, it still remains a real challenge. Moreover, the issue is even more complex as full texts are not always freely available. Then, using only partial information to annotate these documents is promising but remains a very ambitious issue. METHODS: We propose two classification methods: a k-nearest neighbours (kNN)-based approach and an explicit semantic analysis (ESA)-based approach. Although the kNN-based approach is widely used in text classification, it needs to be improved to perform well in this specific classification problem which deals with partial information. Compared to existing kNN-based methods, our method uses classical Machine Learning (ML) algorithms for ranking the labels. Additional features are also investigated in order to improve the classifiers' performance. In addition, the combination of several learning algorithms with various techniques for fixing the number of relevant topics is performed. On the other hand, ESA seems promising for this classification task as it yielded interesting results in related issues, such as semantic relatedness computation between texts and text classification. Unlike existing works, which use ESA for enriching the bag-of-words approach with additional knowledge-based features, our ESA-based method builds a standalone classifier. Furthermore, we investigate if the results of this method could be useful as a complementary feature of our kNN-based approach. RESULTS: Experimental evaluations performed on large standard annotated datasets, provided by the BioASQ organizers, show that the kNN-based method with the Random Forest learning algorithm achieves good performances compared with the current state-of-the-art methods, reaching a competitive f-measure of 0.55 % while the ESA-based approach surprisingly yielded unsatisfactory results. CONCLUSIONS: We have proposed simple classification methods suitable to annotate textual documents using only partial information. They are therefore adequate for large multi-label classification and particularly in the biomedical domain. Thus, our work contributes to the extraction of relevant information from unstructured documents in order to facilitate their automated processing. Consequently, it could be used for various purposes, including document indexing, information retrieval, etc.

Semantic Indexing of Medical Learning Objects: Medical Students' Usage of a Semantic Network.

BACKGROUND: The Semantically Annotated Media (SAM) project aims to provide a flexible platform for searching, browsing, and indexing medical learning objects (MLOs) based on a semantic network derived from established classification systems. Primarily, SAM supports the Aachen emedia skills lab, but SAM is ready for indexing distributed content and the Simple Knowledge Organizing System standard provides a means for easily upgrading or even exchanging SAM's semantic network. There is a lack of research addressing the usability of MLO indexes or search portals like SAM and the user behavior with such platforms. OBJECTIVE: The purpose of this study was to assess the usability of SAM by investigating characteristic user behavior of medical students accessing MLOs via SAM. METHODS: In this study, we chose a mixed-methods approach. Lean usability testing was combined with usability inspection by having the participants complete four typical usage scenarios before filling out a questionnaire. The questionnaire was based on the IsoMetrics usability inventory. Direct user interaction with SAM (mouse clicks and pages accessed) was logged. RESULTS: The study analyzed the typical usage patterns and habits of students using a semantic network for accessing MLOs. Four scenarios capturing characteristics of typical tasks to be solved by using SAM yielded high ratings of usability items and showed good results concerning the consistency of indexing by different users. Long-tail phenomena emerge as they are typical for a collaborative Web 2.0 platform. Suitable but nonetheless rarely used keywords were assigned to MLOs by some users. CONCLUSIONS: It is possible to develop a Web-based tool with high usability and acceptance for indexing and retrieval of MLOs. SAM can be applied to indexing multicentered repositories of MLOs collaboratively.

A unified statistical approach to non-negative matrix factorization and probabilistic latent semantic indexing.

Non-negative matrix factorization (NMF) is a powerful machine learning method for decomposing a high-dimensional nonnegative matrix V into the product of two nonnegative matrices, W and H, such that V ∼ W H. It has been shown to have a parts-based, sparse representation of the data. NMF has been successfully applied in a variety of areas such as natural language processing, neuroscience, information retrieval, image processing, speech recognition and computational biology for the analysis and interpretation of large-scale data. There has also been simultaneous development of a related statistical latent class modeling approach, namely, probabilistic latent semantic indexing (PLSI), for analyzing and interpreting co-occurrence count data arising in natural language processing. In this paper, we present a generalized statistical approach to NMF and PLSI based on Renyi's divergence between two non-negative matrices, stemming from the Poisson likelihood. Our approach unifies various competing models and provides a unique theoretical framework for these methods. We propose a unified algorithm for NMF and provide a rigorous proof of monotonicity of multiplicative updates for W and H. In addition, we generalize the relationship between NMF and PLSI within this framework. We demonstrate the applicability and utility of our approach as well as its superior performance relative to existing methods using real-life and simulated document clustering data.

Effective use of latent semantic indexing and computational linguistics in biological and biomedical applications.

Text mining is rapidly becoming an essential technique for the annotation and analysis of large biological data sets. Biomedical literature currently increases at a rate of several thousand papers per week, making automated information retrieval methods the only feasible method of managing this expanding corpus. With the increasing prevalence of open-access journals and constant growth of publicly-available repositories of biomedical literature, literature mining has become much more effective with respect to the extraction of biomedically-relevant data. In recent years, text mining of popular databases such as MEDLINE has evolved from basic term-searches to more sophisticated natural language processing techniques, indexing and retrieval methods, structural analysis and integration of literature with associated metadata. In this review, we will focus on Latent Semantic Indexing (LSI), a computational linguistics technique increasingly used for a variety of biological purposes. It is noted for its ability to consistently outperform benchmark Boolean text searches and co-occurrence models at information retrieval and its power to extract indirect relationships within a data set. LSI has been used successfully to formulate new hypotheses, generate novel connections from existing data, and validate empirical data.

Gestural cue analysis in automated semantic miscommunication annotation.

The automated annotation of conversational video by semantic miscommunication labels is a challenging topic. Although miscommunications are often obvious to the speakers as well as the observers, it is difficult for machines to detect them from the low-level features. We investigate the utility of gestural cues in this paper among various non-verbal features. Compared with gesture recognition tasks in human-computer interaction, this process is difficult due to the lack of understanding on which cues contribute to miscommunications and the implicitness of gestures. Nine simple gestural features are taken from gesture data, and both simple and complex classifiers are constructed using machine learning. The experimental results suggest that there is no single gestural feature that can predict or explain the occurrence of semantic miscommunication in our setting.

Using linear algebra for protein structural comparison and classification.

In this article, we describe a novel methodology to extract semantic characteristics from protein structures using linear algebra in order to compose structural signature vectors which may be used efficiently to compare and classify protein structures into fold families. These signatures are built from the pattern of hydrophobic intrachain interactions using Singular Value Decomposition (SVD) and Latent Semantic Indexing (LSI) techniques. Considering proteins as documents and contacts as terms, we have built a retrieval system which is able to find conserved contacts in samples of myoglobin fold family and to retrieve these proteins among proteins of varied folds with precision of up to 80%. The classifier is a web tool available at our laboratory website. Users can search for similar chains from a specific PDB, view and compare their contact maps and browse their structures using a JMol plug-in.

Using linear algebra for protein structural comparison and classification

In this article, we describe a novel methodology to extract semantic characteristics from protein structures using linear algebra in order to compose structural signature vectors which may be used efficiently to compare and classify protein structures into fold families. These signatures are built from the pattern of hydrophobic intrachain interactions using Singular Value Decomposition (SVD) and Latent Semantic Indexing (LSI) techniques. Considering proteins as documents and contacts as terms, we have built a retrieval system which is able to find conserved contacts in samples of myoglobin fold family and to retrieve these proteins among proteins of varied folds with precision of up to 80 percent. The classifier is a web tool available at our laboratory website. Users can search for similar chains from a specific PDB, view and compare their contact maps and browse their structures using a JMol plug-in.