Abstract
Precision medicine aims to provide personalized treatments based on individual patient profiles. One critical step towards precision medicine is leveraging knowledge derived from biomedical publications—a tremendous literature resource presenting the latest scientific discoveries on genes, mutations and diseases. Biomedical natural language processing (BioNLP) plays a vital role in supporting automation of this process. BioCreative VI Track 4 brings community effort to the task of automatically identifying and extracting protein–protein interactions (PPi) affected by mutations (PPIm), important in the precision medicine context for capturing individual genotype variation related to disease.We present the READ-BioMed team’s approach to identifying PPIm-related publications and to extracting specific PPIm information from those publications in the context of the BioCreative VI PPIm track. We observe that current BioNLP tools are insufficient to recognise entities for these two tasks; the best existing mutation recognition tool achieves only 55% recall in the document triage training set, while relation extraction performance is limited by the low recall performance of gene entity recognition. We develop the models accordingly: for document triage, we develop term lists capturing interactions and mutations to complement BioNLP tools, and select effective features via a feature contribution study, whereas an ensemble of BioNLP tools is employed for relation extraction.Our best document triage model achieves an F-score of 66.77% while our best model for relation extraction achieved an F-score of 35.09% over the final (updated post-task) test set. Impacting the document triage task, the characteristics of mutations are statistically different in the training and testing sets. While a vital new direction for biomedical text mining research, this early attempt to tackle the problem of identifying genetic variation of substantial biological significance highlights the importance of representative training data and the cascading impact of tool limitations in a modular system.
Highlights
Precision medicine is an emerging field [1], aiming to provide specialized medical treatments on the basis of individual patient characteristics, including their genotype, phenotype and other diagnostics [2]
In the official results of this shared task presented at the BioCreative VI workshop, our best document triage model achieved over 88% recall, achieving the third highest recall amongst 22 submissions, while our best model for relation extraction achieved a Micro F1-score of 37.17%, ranking second amongst six submissions just behind the top team at 37.29%
We find mutation characteristics have different impacts on both tasks: for document triage, the model performance is dramatically decreased due to the significantly different mutation characteristics between the training and testing set; in contrast, for relation extraction, recognising mutations facilitates the extraction of PPI affected by mutations (PPIm) relations
Summary
Precision medicine is an emerging field [1], aiming to provide specialized medical treatments on the basis of individual patient characteristics, including their genotype, phenotype and other diagnostics [2]. PubMed, the primary biomedical literature database, contains over 28 million biomedical publications (https://www.ncbi.nlm.nih.gov/pubmed/). This literature represents a critical information source for precision medicine, but the vast quantities of unstructured text make it challenging to identify and navigate relevant evidence. Two primary BioNLP tasks relevant to precision medicine are named entity recognition, e.g. as applied to recognize mentions of mutations in articles [4] and relation extraction, e.g. to identify interactions, such as protein–protein interactions (PPI), between biological entities described in papers [5]. In the context of precision medicine, identification and extraction of PPI affected by mutations (PPIm) described in the literature [8] supports synthesis and, in turn, deeper understanding of the biological impacts of genetic variation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
