The study aims at developing a neural network model to improve the performance of Human Phenotype Ontology (HPO) concept recognition tools. We used the terms, definitions, and comments about the phenotypic concepts in the HPO database to train our model. The document to be analyzed is first split into sentences and annotated with a base method to generate candidate concepts. The sentences, along with the candidate concepts, are then fed into the pre-trained model for re-ranking. Our model comprises the pre-trained BlueBERT and a feature selection module, followed by a contrastive loss. We re-ranked the results generated by three robust HPO annotation tools and compared the performance against most of the existing approaches. The experimental results show that our model can improve the performance of the existing methods. Significantly, it boosted 3.0% and 5.6% in F1 score on the two evaluated datasets compared with the base methods. It removed more than 80% of the false positives predicted by the base methods, resulting in up to 18% improvement in precision. Our model utilizes the descriptive data in the ontology and the contextual information in the sentences for re-ranking. The results indicate that the additional information and the re-ranking model can significantly enhance the precision of HPO concept recognition compared with the base method.

MOTIVATION: Automatic phenotype concept recognition from unstructured text remains a challenging task in biomedical text mining research. Previous works that address the task typically use dictionary-based matching methods, which can achieve high precision but suffer from lower recall. Recently, machine learning-based methods have been proposed to identify biomedical concepts, which can recognize more unseen concept synonyms by automatic feature learning. However, most methods require large corpora of manually annotated data for model training, which is difficult to obtain due to the high cost of human annotation. RESULTS: In this article, we propose PhenoTagger, a hybrid method that combines both dictionary and machine learning-based methods to recognize Human Phenotype Ontology (HPO) concepts in unstructured biomedical text. We first use all concepts and synonyms in HPO to construct a dictionary, which is then used to automatically build a distantly supervised training dataset for machine learning. Next, a cutting-edge deep learning model is trained to classify each candidate phrase (n-gram from input sentence) into a corresponding concept label. Finally, the dictionary and machine learning-based prediction results are combined for improved performance. Our method is validated with two HPO corpora, and the results show that PhenoTagger compares favorably to previous methods. In addition, to demonstrate the generalizability of our method, we retrained PhenoTagger using the disease ontology MEDIC for disease concept recognition to investigate the effect of training on different ontologies. Experimental results on the NCBI disease corpus show that PhenoTagger without requiring manually annotated training data achieves competitive performance as compared with state-of-the-art supervised methods. AVAILABILITYAND IMPLEMENTATION: The source code, API information and data for PhenoTagger are freely available at https://github.com/ncbi-nlp/PhenoTagger. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The CEBS data repository is being developed to promote a systems biology approach to understand the biological effects of environmental stressors. CEBS will house data from multiple gene expression platforms (transcriptomics), protein expression and protein-protein interaction (proteomics), and changes in low molecular weight metabolite levels (metabolomics) aligned by their detailed toxicological context. The system will accommodate extensive complex querying in a user-friendly manner. CEBS will store toxicological contexts including the study design details, treatment protocols, animal characteristics and conventional toxicological endpoints such as histopathology findings and clinical chemistry measures. All of these data types can be integrated in a seamless fashion to enable data query and analysis in a biologically meaningful manner. RESULTS: An object model, the SysBio-OM (Xirasagar et al., 2004) has been designed to facilitate the integration of microarray gene expression, proteomics and metabolomics data in the CEBS database system. We now report SysTox-OM as an open source systems toxicology model designed to integrate toxicological context into gene expression experiments. The SysTox-OM model is comprehensive and leverages other open source efforts, namely, the Standard for Exchange of Nonclinical Data (http://www.cdisc.org/models/send/v2/index.html) which is a data standard for capturing toxicological information for animal studies and Clinical Data Interchange Standards Consortium (http://www.cdisc.org/models/sdtm/index.html) that serves as a standard for the exchange of clinical data. Such standardization increases the accuracy of data mining, interpretation and exchange. The open source SysTox-OM model, which can be implemented on various software platforms, is presented here. AVAILABILITY: A universal modeling language (UML) depiction of the entire SysTox-OM is available at http://cebs.niehs.nih.gov and the Rational Rose object model package is distributed under an open source license that permits unrestricted academic and commercial use and is available at http://cebs.niehs.nih.gov/cebsdownloads. Currently, the public toxicological data in CEBS can be queried via a web application based on the SysTox-OM at http://cebs.niehs.nih.gov CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: To promote a systems biology approach to understanding the biological effects of environmental stressors, the Chemical Effects in Biological Systems (CEBS) knowledge base is being developed to house data from multiple complex data streams in a systems friendly manner that will accommodate extensive querying from users. Unified data representation via a single object model will greatly aid in integrating data storage and management, and facilitate reuse of software to analyze and display data resulting from diverse differential expression or differential profile technologies. Data streams include, but are not limited to, gene expression analysis (transcriptomics), protein expression and protein-protein interaction analysis (proteomics) and changes in low molecular weight metabolite levels (metabolomics). RESULTS: To enable the integration of microarray gene expression, proteomics and metabolomics data in the CEBS system, we designed an object model, Systems Biology Object Model (SysBio-OM). The model is comprehensive and leverages other open source efforts, namely the MicroArray Gene Expression Object Model (MAGE-OM) and the Proteomics Experiment Data Repository (PEDRo) object model. SysBio-OM is designed by extending MAGE-OM to represent protein expression data elements (including those from PEDRo), protein-protein interaction and metabolomics data. SysBio-OM promotes the standardization of data representation and data quality by facilitating the capture of the minimum annotation required for an experiment. Such standardization refines the accuracy of data mining and interpretation. The open source SysBio-OM model, which can be implemented on varied computing platforms is presented here. AVAILABILITY: A universal modeling language depiction of the entire SysBio-OM is available at http://cebs.niehs.nih.gov/SysBioOM/. The Rational Rose object model package is distributed under an open source license that permits unrestricted academic and commercial use and is available at http://cebs.niehs.nih.gov/cebsdownloads. The database and interface are being built to implement the model and will be available for public use at http://cebs.niehs.nih.gov.