This article reports on a detailed investigation of PubMed users' needs and behavior as a step toward improving biomedical information retrieval. PubMed is providing free service to researchers with access to more than 19 million citations for biomedical articles from MEDLINE and life science journals. It is accessed by millions of users each day. Efficient search tools are crucial for biomedical researchers to keep abreast of the biomedical literature relating to their own research. This study provides insight into PubMed users' needs and their behavior. This investigation was conducted through the analysis of one month of log data, consisting of more than 23 million user sessions and more than 58 million user queries. Multiple aspects of users' interactions with PubMed are characterized in detail with evidence from these logs. Despite having many features in common with general Web searches, biomedical information searches have unique characteristics that are made evident in this study. PubMed users are more persistent in seeking information and they reformulate queries often. The three most frequent types of search are search by author name, search by gene/protein, and search by disease. Use of abbreviation in queries is very frequent. Factors such as result set size influence users' decisions. Analysis of characteristics such as these plays a critical role in identifying users' information needs and their search habits. In turn, such an analysis also provides useful insight for improving biomedical information retrieval.Database URL:http://www.ncbi.nlm.nih.gov/PubMed.

The automatic extraction of chemical information from text requires the recognition of chemical entity mentions as one of its key steps. When developing supervised named entity recognition (NER) systems, the availability of a large, manually annotated text corpus is desirable. Furthermore, large corpora permit the robust evaluation and comparison of different approaches that detect chemicals in documents. We present the CHEMDNER corpus, a collection of 10,000 PubMed abstracts that contain a total of 84,355 chemical entity mentions labeled manually by expert chemistry literature curators, following annotation guidelines specifically defined for this task. The abstracts of the CHEMDNER corpus were selected to be representative for all major chemical disciplines. Each of the chemical entity mentions was manually labeled according to its structure-associated chemical entity mention (SACEM) class: abbreviation, family, formula, identifier, multiple, systematic and trivial. The difficulty and consistency of tagging chemicals in text was measured using an agreement study between annotators, obtaining a percentage agreement of 91. For a subset of the CHEMDNER corpus (the test set of 3,000 abstracts) we provide not only the Gold Standard manual annotations, but also mentions automatically detected by the 26 teams that participated in the BioCreative IV CHEMDNER chemical mention recognition task. In addition, we release the CHEMDNER silver standard corpus of automatically extracted mentions from 17,000 randomly selected PubMed abstracts. A version of the CHEMDNER corpus in the BioC format has been generated as well. We propose a standard for required minimum information about entity annotations for the construction of domain specific corpora on chemical and drug entities. The CHEMDNER corpus and annotation guidelines are available at: http://www.biocreative.org/resources/biocreative-iv/chemdner-corpus/.

Chemical compounds and drugs are an important class of entities in biomedical research with great potential in a wide range of applications, including clinical medicine. Locating chemical named entities in the literature is a useful step in chemical text mining pipelines for identifying the chemical mentions, their properties, and their relationships as discussed in the literature. We introduce the tmChem system, a chemical named entity recognizer created by combining two independent machine learning models in an ensemble. We use the corpus released as part of the recent CHEMDNER task to develop and evaluate tmChem, achieving a micro-averaged f-measure of 0.8739 on the CEM subtask (mention-level evaluation) and 0.8745 f-measure on the CDI subtask (abstract-level evaluation). We also report a high-recall combination (0.9212 for CEM and 0.9224 for CDI). tmChem achieved the highest f-measure reported in the CHEMDNER task for the CEM subtask, and the high recall variant achieved the highest recall on both the CEM and CDI tasks. We report that tmChem is a state-of-the-art tool for chemical named entity recognition and that performance for chemical named entity recognition has now tied (or exceeded) the performance previously reported for genes and diseases. Future research should focus on tighter integration between the named entity recognition and normalization steps for improved performance. The source code and a trained model for both models of tmChem is available at: http://www.ncbi.nlm.nih.gov/CBBresearch/Lu/Demo/tmChem. The results of running tmChem (Model 2) on PubMed are available in PubTator: http://www.ncbi.nlm.nih.gov/CBBresearch/Lu/Demo/PubTator.

Motivated by the high cost of human curation of biological databases, there is an increasing interest in using computational approaches to assist human curators and accelerate the manual curation process. Towards the goal of cataloging drug indications from FDA drug labels, we recently developed LabeledIn, a human-curated drug indication resource for 250 clinical drugs. Its development required over 40 h of human effort across 20 weeks, despite using well-defined annotation guidelines. In this study, we aim to investigate the feasibility of scaling drug indication annotation through a crowdsourcing technique where an unknown network of workers can be recruited through the technical environment of Amazon Mechanical Turk (MTurk). To translate the expert-curation task of cataloging indications into human intelligence tasks (HITs) suitable for the average workers on MTurk, we first simplify the complex task such that each HIT only involves a worker making a binary judgment of whether a highlighted disease, in context of a given drug label, is an indication. In addition, this study is novel in the crowdsourcing interface design where the annotation guidelines are encoded into user options. For evaluation, we assess the ability of our proposed method to achieve high-quality annotations in a time-efficient and cost-effective manner. We posted over 3000 HITs drawn from 706 drug labels on MTurk. Within 8 h of posting, we collected 18 775 judgments from 74 workers, and achieved an aggregated accuracy of 96% on 450 control HITs (where gold-standard answers are known), at a cost of $1.75 per drug label. On the basis of these results, we conclude that our crowdsourcing approach not only results in significant cost and time saving, but also leads to accuracy comparable to that of domain experts.

BioC is a new simple XML format for sharing biomedical text and annotations and libraries to read and write that format. This promotes the development of interoperable tools for natural language processing (NLP) of biomedical text. The interoperability track at the BioCreative IV workshop featured contributions using or highlighting the BioC format. These contributions included additional implementations of BioC, many new corpora in the format, biomedical NLP tools consuming and producing the format and online services using the format. The ease of use, broad support and rapidly growing number of tools demonstrate the need for and value of the BioC format. Database URL: http://bioc.sourceforge.net/.

Manual curation of data from the biomedical literature is a rate-limiting factor for many expert curated databases. Despite the continuing advances in biomedical text mining and the pressing needs of biocurators for better tools, few existing text-mining tools have been successfully integrated into production literature curation systems such as those used by the expert curated databases. To close this gap and better understand all aspects of literature curation, we invited submissions of written descriptions of curation workflows from expert curated databases for the BioCreative 2012 Workshop Track II. We received seven qualified contributions, primarily from model organism databases. Based on these descriptions, we identified commonalities and differences across the workflows, the common ontologies and controlled vocabularies used and the current and desired uses of text mining for biocuration. Compared to a survey done in 2009, our 2012 results show that many more databases are now using text mining in parts of their curation workflows. In addition, the workshop participants identified text-mining aids for finding gene names and symbols (gene indexing), prioritization of documents for curation (document triage) and ontology concept assignment as those most desired by the biocurators. DATABASE URL: http://www.biocreative.org/tasks/bc-workshop-2012/workflow/.

Community-run, formal evaluations and manually annotated text corpora are critically important for advancing biomedical text-mining research. Recently in BioCreative V, a new challenge was organized for the tasks of disease named entity recognition (DNER) and chemical-induced disease (CID) relation extraction. Given the nature of both tasks, a test collection is required to contain both disease/chemical annotations and relation annotations in the same set of articles. Despite previous efforts in biomedical corpus construction, none was found to be sufficient for the task. Thus, we developed our own corpus called BC5CDR during the challenge by inviting a team of Medical Subject Headings (MeSH) indexers for disease/chemical entity annotation and Comparative Toxicogenomics Database (CTD) curators for CID relation annotation. To ensure high annotation quality and productivity, detailed annotation guidelines and automatic annotation tools were provided. The resulting BC5CDR corpus consists of 1500 PubMed articles with 4409 annotated chemicals, 5818 diseases and 3116 chemical-disease interactions. Each entity annotation includes both the mention text spans and normalized concept identifiers, using MeSH as the controlled vocabulary. To ensure accuracy, the entities were first captured independently by two annotators followed by a consensus annotation: The average inter-annotator agreement (IAA) scores were 87.49% and 96.05% for the disease and chemicals, respectively, in the test set according to the Jaccard similarity coefficient. Our corpus was successfully used for the BioCreative V challenge tasks and should serve as a valuable resource for the text-mining research community.Database URL: http://www.biocreative.org/tasks/biocreative-v/track-3-cdr/.

The biomedical literature provides a rich source of knowledge such as protein-protein interactions (PPIs), drug-drug interactions (DDIs) and chemical-protein interactions (CPIs). Biomedical relation extraction aims to automatically extract biomedical relations from biomedical text for various biomedical research. State-of-the-art methods for biomedical relation extraction are primarily based on supervised machine learning and therefore depend on (sufficient) labeled data. However, creating large sets of training data is prohibitively expensive and labor-intensive, especially so in biomedicine as domain knowledge is required. In contrast, there is a large amount of unlabeled biomedical text available in PubMed. Hence, computational methods capable of employing unlabeled data to reduce the burden of manual annotation are of particular interest in biomedical relation extraction. We present a novel semi-supervised approach based on variational autoencoder (VAE) for biomedical relation extraction. Our model consists of the following three parts, a classifier, an encoder and a decoder. The classifier is implemented using multi-layer convolutional neural networks (CNNs), and the encoder and decoder are implemented using both bidirectional long short-term memory networks (Bi-LSTMs) and CNNs, respectively. The semi-supervised mechanism allows our model to learn features from both the labeled and unlabeled data. We evaluate our method on multiple public PPI, DDI and CPI corpora. Experimental results show that our method effectively exploits the unlabeled data to improve the performance and reduce the dependence on labeled data. To our best knowledge, this is the first semi-supervised VAE-based method for (biomedical) relation extraction. Our results suggest that exploiting such unlabeled data can be greatly beneficial to improved performance in various biomedical relation extraction, especially when only limited labeled data (e.g. 2000 samples or less) is available in such tasks.

In biomedicine, key concepts are often expressed by multiple words (e.g., 'zinc finger protein'). Previous work has shown treating a sequence of words as a meaningful unit, where applicable, is not only important for human understanding but also beneficial for automatic information seeking. Here we present a collection of PubMed® Phrases that are beneficial for information retrieval and human comprehension. We define these phrases as coherent chunks that are logically connected. To collect the phrase set, we apply the hypergeometric test to detect segments of consecutive terms that are likely to appear together in PubMed. These text segments are then filtered using the BM25 ranking function to ensure that they are beneficial from an information retrieval perspective. Thus, we obtain a set of 705,915 PubMed Phrases. We evaluate the quality of the set by investigating PubMed user click data and manually annotating a sample of 500 randomly selected noun phrases. We also analyze and discuss the usage of these PubMed Phrases in literature search.

PubMed® is a search engine providing access to a collection of over 27 million biomedical bibliographic records as of 2017. PubMed processes millions of queries a day, and understanding these queries is one of the main building blocks for successful information retrieval. In this work, we present Field Sensor, a domain-specific tool for understanding the composition and predicting the user intent of PubMed queries. Given a query, the Field Sensor infers a field for each token or sequence of tokens in a query in multi-step process that includes syntactic chunking, rule-based tagging and probabilistic field prediction. In this work, the fields of interest are those associated with (meta-)data elements of each PubMed record such as article title, abstract, author name(s), journal title, volume, issue, page and date. We evaluate the accuracy of our algorithm on a human-annotated corpus of 10 000 PubMed queries, as well as a new machine-annotated set of 103 000 PubMed queries. The Field Sensor achieves an accuracy of 93 and 91% on the two corresponding corpora and finds that nearly half of all searches are navigational (e.g. author searches, article title searches etc.) and half are informational (e.g. topical searches). The Field Sensor has been integrated into PubMed since June 2017 to detect informational queries for which results sorted by relevance can be suggested as an alternative to those sorted by the default date sort. In addition, the composition of PubMed queries as computed by the Field Sensor proves to be essential for understanding how users query PubMed.

Automatically identifying chemical and drug names in scientific publications advances information access for this important class of entities in a variety of biomedical disciplines by enabling improved retrieval and linkage to related concepts. While current methods for tagging chemical entities were developed for the article title and abstract, their performance in the full article text is substantially lower. However, the full text frequently contains more detailed chemical information, such as the properties of chemical compounds, their biological effects and interactions with diseases, genes and other chemicals. We therefore present the NLM-Chem corpus, a full-text resource to support the development and evaluation of automated chemical entity taggers. The NLM-Chem corpus consists of 150 full-text articles, doubly annotated by ten expert NLM indexers, with ~5000 unique chemical name annotations, mapped to ~2000 MeSH identifiers. We also describe a substantially improved chemical entity tagger, with automated annotations for all of PubMed and PMC freely accessible through the PubTator web-based interface and API. The NLM-Chem corpus is freely available.

Semantic textual similarity (STS) measures the degree of relatedness between sentence pairs. The Open Health Natural Language Processing (OHNLP) Consortium released an expertly annotated STS data set and called for the National Natural Language Processing Clinical Challenges. This work describes our entry, an ensemble model that leverages a range of deep learning (DL) models. Our team from the National Library of Medicine obtained a Pearson correlation of 0.8967 in an official test set during 2019 National Natural Language Processing Clinical Challenges/Open Health Natural Language Processing shared task and achieved a second rank.

The rapid growth of biomedical literature poses a significant challenge for curation and interpretation. This has become more evident during the COVID-19 pandemic. LitCovid, a literature database of COVID-19 related papers in PubMed, has accumulated over 200,000 articles with millions of accesses. Approximately 10,000 new articles are added to LitCovid every month. A main curation task in LitCovid is topic annotation where an article is assigned with up to eight topics, e.g., Treatment and Diagnosis. The annotated topics have been widely used both in LitCovid (e.g., accounting for ∼18% of total uses) and downstream studies such as network generation. However, it has been a primary curation bottleneck due to the nature of the task and the rapid literature growth. This study proposes LITMC-BERT, a transformer-based multi-label classification method in biomedical literature. It uses a shared transformer backbone for all the labels while also captures label-specific features and the correlations between label pairs. We compare LITMC-BERT with three baseline models on two datasets. Its micro-F1 and instance-based F1 are 5% and 4% higher than the current best results, respectively, and only requires ∼18% of the inference time than the Binary BERT baseline. The related datasets and models are available via https://github.com/ncbi/ml-transformer.

Pretrained language models such as Bidirectional Encoder Representations from Transformers (BERT) have achieved state-of-the-art performance in natural language processing (NLP) tasks. Recently, BERT has been adapted to the biomedical domain. Despite the effectiveness, these models have hundreds of millions of parameters and are computationally expensive when applied to large-scale NLP applications. We hypothesized that the number of parameters of the original BERT can be dramatically reduced with minor impact on performance. In this study, we present Bioformer, a compact BERT model for biomedical text mining. We pretrained two Bioformer models (named Bioformer8L and Bioformer16L) which reduced the model size by 60% compared to BERTBase. Bioformer uses a biomedical vocabulary and was pre-trained from scratch on PubMed abstracts and PubMed Central full-text articles. We thoroughly evaluated the performance of Bioformer as well as existing biomedical BERT models including BioBERT and PubMedBERT on 15 benchmark datasets of four different biomedical NLP tasks: named entity recognition, relation extraction, question answering and document classification. The results show that with 60% fewer parameters, Bioformer16L is only 0.1% less accurate than PubMedBERT while Bioformer8L is 0.9% less accurate than PubMedBERT. Both Bioformer16L and Bioformer8L outperformed BioBERTBase-v1.1. In addition, Bioformer16L and Bioformer8L are 2-3 fold as fast as PubMedBERT/BioBERTBase-v1.1. Bioformer has been successfully deployed to PubTator Central providing gene annotations over 35 million PubMed abstracts and 5 million PubMed Central full-text articles. We make Bioformer publicly available via https://github.com/WGLab/bioformer, including pre-trained models, datasets, and instructions for downstream use.

In addition to maintaining the GenBank® nucleic acid sequence database, the National Center for Biotechnology Information (NCBI) provides analysis and retrieval resources for the data in GenBank and other biological data made available through the NCBI Web site. NCBI resources include Entrez, the Entrez Programming Utilities, MyNCBI, PubMed, PubMed Central (PMC), Entrez Gene, the NCBI Taxonomy Browser, BLAST, BLAST Link (BLink), Primer-BLAST, COBALT, Electronic PCR, OrfFinder, Splign, ProSplign, RefSeq, UniGene, HomoloGene, ProtEST, dbMHC, dbSNP, dbVar, Epigenomics, Cancer Chromosomes, Entrez Genomes and related tools, the Map Viewer, Model Maker, Evidence Viewer, Trace Archive, Sequence Read Archive, Retroviral Genotyping Tools, HIV-1/Human Protein Interaction Database, Gene Expression Omnibus (GEO), Entrez Probe, GENSAT, Online Mendelian Inheritance in Man (OMIM), Online Mendelian Inheritance in Animals (OMIA), the Molecular Modeling Database (MMDB), the Conserved Domain Database (CDD), the Conserved Domain Architecture Retrieval Tool (CDART), IBIS, Biosystems, Peptidome, OMSSA, Protein Clusters and the PubChem suite of small molecule databases. Augmenting many of the Web applications are custom implementations of the BLAST program optimized to search specialized data sets. All of these resources can be accessed through the NCBI home page at www.ncbi.nlm.nih.gov.

The role of pathological response in long-term outcome is still unclear in cervical cancer patients treated with neoadjuvant chemotherapy (NACT) in China. This study aimed to investigate the effect of optimal pathologic response (OPR) on survival in the patients treated with NACT and radical hysterectomy. First, 853 patients with stage IB2-IIB cervical cancer were included in a retrospective analysis; a Cox proportional hazards model was used to investigate the relationship between pathological response and disease-free survival (DFS). In the retrospective database, 64 (7.5%) patients were found to have achieved an OPR (residual disease <3 mm stromal invasion); in the multivariate Cox model, the risk of death was much greater in the non-OPR group than in the OPR group (HR, 2.61; 95%CI, 1.06 to 6.45; P = 0.037). Next, the role of OPR was also evaluated in a prospective cohort of 603 patients with cervical cancer. In the prospective cohort, 56 (9.3%) patients were found to have achieved an OPR; the log-rank tests showed that the risk of recurrence was higher in the non-OPR patients than in the OPR group (P = 0.05). After combined analysis, OPR in cervical cancer was found to be an independent prognostic factor for DFS.

Identifying relevant papers from the literature is a common task in biocuration. Most current biomedical literature search systems primarily rely on matching user keywords. Semantic search, on the other hand, seeks to improve search accuracy by understanding the entities and contextual relations in user keywords. However, past research has mostly focused on semantically identifying biological entities (e.g. chemicals, diseases and genes) with little effort on discovering semantic relations. In this work, we aim to discover biomedical semantic relations in PubMed queries in an automated and unsupervised fashion. Specifically, we focus on extracting and understanding the contextual information (or context patterns) that is used by PubMed users to represent semantic relations between entities such as 'CHEMICAL-1 compared to CHEMICAL-2' With the advances in automatic named entity recognition, we first tag entities in PubMed queries and then use tagged entities as knowledge to recognize pattern semantics. More specifically, we transform PubMed queries into context patterns involving participating entities, which are subsequently projected to latent topics via latent semantic analysis (LSA) to avoid the data sparseness and specificity issues. Finally, we mine semantically similar contextual patterns or semantic relations based on LSA topic distributions. Our two separate evaluation experiments of chemical-chemical (CC) and chemical-disease (CD) relations show that the proposed approach significantly outperforms a baseline method, which simply measures pattern semantics by similarity in participating entities. The highest performance achieved by our approach is nearly 0.9 and 0.85 respectively for the CC and CD task when compared against the ground truth in terms of normalized discounted cumulative gain (nDCG), a standard measure of ranking quality. These results suggest that our approach can effectively identify and return related semantic patterns in a ranked order covering diverse bio-entity relations. To assess the potential utility of our automated top-ranked patterns of a given relation in semantic search, we performed a pilot study on frequently sought semantic relations in PubMed and observed improved literature retrieval effectiveness based on post-hoc human relevance evaluation. Further investigation in larger tests and in real-world scenarios is warranted.

MeSH indexing is the task of assigning relevant MeSH terms based on a manual reading of scholarly publications by human indexers. The task is highly important for improving literature retrieval and many other scientific investigations in biomedical research. Unfortunately, given its manual nature, the process of MeSH indexing is both time-consuming (new articles are not immediately indexed until 2 or 3 months later) and costly (approximately ten dollars per article). In response, automatic indexing by computers has been previously proposed and attempted but remains challenging. In order to advance the state of the art in automatic MeSH indexing, a community-wide shared task called BioASQ was recently organized.

Recently, advanced text-mining techniques have been shown to speed up manual data curation by providing human annotators with automated pre-annotations generated by rules or machine learning models. Due to the limited training data available, however, current annotation systems primarily focus only on common concept types such as genes or diseases. To support annotating a wide variety of biological concepts with or without pre-existing training data, we developed ezTag, a web-based annotation tool that allows curators to perform annotation and provide training data with humans in the loop. ezTag supports both abstracts in PubMed and full-text articles in PubMed Central. It also provides lexicon-based concept tagging as well as the state-of-the-art pre-trained taggers such as TaggerOne, GNormPlus and tmVar. ezTag is freely available at http://eztag.bioqrator.org.

High-throughput experiments and bioinformatics techniques are creating an exploding volume of data that are becoming overwhelming to keep track of for biologists and researchers who need to access, analyze and process existing data. Much of the available data are being deposited in specialized databases, such as the Gene Expression Omnibus (GEO) for microarrays or the Protein Data Bank (PDB) for protein structures and coordinates. Data sets are also being described by their authors in publications archived in literature databases such as MEDLINE and PubMed Central. Currently, the curation of links between biological databases and the literature mainly relies on manual labour, which makes it a time-consuming and daunting task. Herein, we analysed the current state of link curation between GEO, PDB and MEDLINE. We found that the link curation is heterogeneous depending on the sources and databases involved, and that overlap between sources is low, <50% for PDB and GEO. Furthermore, we showed that text-mining tools can automatically provide valuable evidence to help curators broaden the scope of articles and database entries that they review. As a result, we made recommendations to improve the coverage of curated links, as well as the consistency of information available from different databases while maintaining high-quality curation. Database URLs: http://www.ncbi.nlm.nih.gov/PubMed, http://www.ncbi.nlm.nih.gov/geo/, http://www.rcsb.org/pdb/