Phenebank: Processed Medline Abstracts

<strong>The PheneBank project:</strong> Free text scientific literature has the potential to be an incredibly valuable source of data for uncovering the often hidden relationships between genes, diseases and phenotypes. Phenotypic descriptions cover abnormalities in anatomical structures, processes and behaviours. For example 'growth delay' and 'body weight loss'. Such descriptions form the basis for determining the existence and treatment of a disease but, because of their inherent complexity, have previously received less attention by the text mining community. In recent years, significant effort has been spent by a small number of expert curators to create coding systems for phenotypes (called "ontologies"), such as the Human Phenotype Ontology (HP) and the Mammalian Phenotype Ontology (MP). The PheneBank project proposes to support and speed up curation using terms discovered directly from the literature and to automatically integrate them with such standard ontologies. <br> <br> The project seeks to harness texts for extracting statistically significant associations between phenotypes, diseases and genes. Earlier approaches have suffered from not providing deep semantic representations of the phenotypes they tried to target. Our deep learning-based approach is an attempt to overcome this issue by reducing the uncertainty between textual and ontological forms of phenotypes. Specifically, the model treats multitoken named entities as a single token which allows more reliable handling of multiword expressions. The approach builds on ground breaking research at the European Bininformatics Institute by the PI (Nigel Collier) and the Co-investigator (Damian Smedley, Queen Mary University London), including terminology alignment of phenotypes using pairwise scoring of the conceptual elements that make up the phenotype. https://sites.google.com/site/nhcollier/projects/phenebank <br> <strong>The dataset:</strong> As an output of the PheneBank project, we release the set of 24 million MEDLINE abstracts annotated with 9 classes of entity: Phenotype, Disease, Anatomy, Cell, Cell_line, GPR, Gene_variant, Molecule, and Pathway. The entities have been mapped to five major ontologies: SNOMED, HPO, MeSH, PRO, and FMA. <br> <strong>Processing:</strong> The NER tagging has been done using a BiLSTM-CRF neural model trained on expert-annotated data (to be released for research). The grounding to ontologies relies on semantic embedding of concepts and entities in a unified semantic space. <br> <strong>Data format:</strong> The zip file contains 24359010 .txt files that are classified into 812 directories. Each .txt file is named with a PubMed article ID and contains the corresponding article's abstract and its annotations. Each line starts with a word; for those words that are identified as entities, entity type and mapping information are followed in the same line (tab separated), with the following format: word &lt;TAB&gt; ::: &lt;TAB&gt; entity_type &lt;TAB&gt; entity_concept_ID_1##confidence_score_1 entity_concept_ID_2##confidence_score_2 ... Note that the concepts are sorted according to their mapping confidence scores.**PheneBank项目：** 非结构化科技文本是极具价值的数据来源，可用于挖掘基因、疾病与表型（phenotype）之间常被隐匿的关联。表型描述涵盖解剖结构、生理过程与行为表现的异常情况，例如“发育迟缓”与“体重下降”。此类描述是确定疾病存在与治疗方案的基础，但由于其内在复杂性，此前并未受到文本挖掘领域的足够重视。 近年来，少数专家编目员投入大量精力构建了针对表型的编码系统（即“本体（ontology）”），例如人类表型本体（Human Phenotype Ontology，HP）与哺乳动物表型本体（Mammalian Phenotype Ontology，MP）。PheneBank项目旨在借助直接从文献中提取的术语，辅助并加速表型编目工作，并将这些术语与上述标准本体自动进行整合。 本项目旨在利用文本数据，提取表型、疾病与基因之间具有统计学显著性的关联。此前的相关方法均存在缺陷：无法对目标表型进行深度语义表征。本项目基于深度学习的方法旨在解决这一问题，通过降低表型的文本形式与本体形式之间的不确定性来实现。具体而言，该模型将多Token命名实体视为单个Token，从而能够更可靠地处理多词表达。该方法基于项目首席研究员奈杰尔·科利尔（Nigel Collier）与联合研究员达米安·斯梅德利（Damian Smedley，伦敦玛丽女王大学）在欧洲生物信息研究所开展的开创性研究，其中包括通过对构成表型的概念元素进行两两评分，实现表型的术语对齐。 项目官网：https://sites.google.com/site/nhcollier/projects/phenebank **数据集：** 作为PheneBank项目的产出之一，我们发布了包含2400万条MEDLINE数据库摘要的数据集，这些摘要被标注了9类实体：表型（Phenotype）、疾病（Disease）、解剖结构（Anatomy）、细胞（Cell）、细胞系（Cell_line）、GPR、基因变异体（Gene_variant）、分子（Molecule）与通路（Pathway）。上述实体已被映射至5个主流本体：SNOMED、HPO、MeSH、PRO与FMA。 **数据处理流程：** 本数据集采用基于专家标注数据训练的BiLSTM-CRF神经网络模型完成命名实体识别（Named Entity Recognition，NER）标注（该标注数据将面向科研场景公开）。实体与本体的对齐工作则依赖于将概念与实体映射至统一语义空间的语义嵌入技术。 **数据格式：** 压缩包内包含24359010个.txt格式文件，这些文件被划分为812个目录。每个.txt文件以PubMed文章ID命名，文件内容包含对应文章的摘要及其标注信息。文件内每一行以一个单词开头；对于被识别为实体的单词，该行后续会以制表符分隔的形式附带实体类型与映射信息，格式如下： 单词 <TAB> ::: <TAB> 实体类型 <TAB> 实体概念ID_1##置信度得分_1 实体概念ID_2##置信度得分_2 … 请注意，实体概念将按照映射置信度得分进行排序。