Metabolite Named Entity Recognition: A Hybrid Approach

Kongburan, Wutthipong; Padungweang, Praisan; Krathu, Worarat; Chan, Jonathan H.

doi:10.1007/978-3-319-46687-3_50

Cited by 3 publications

(9 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have described three new DL models, using different word embedding layers, that have both achieved state-ofthe-art performance (F1-score>0.89) for metabolite NER on a manually annotated dataset with OA metabolomics articles. Our DL models (F1-scores of 0.91, 0.90 and 0.89 for TABoLiSTM (BioBERT), TABoLiSTM (BERT) and MetaboListem, respectively) surpass by a considerable margin previous methods (using CRFs) for metabolite NER which achieved maximum F1-scores of 0.78 (8) and 0.87 (9). Our methods, compared to similar methods for chemical entity recognition (17,18,24) and metabolite NER (8,9), have the benefit of having been trained not only on abstracts and titles, but also on full-text paragraphs that are relevant for information retrieval from studies (results and discussion sections).…”

Section: Discussionmentioning

confidence: 84%

“…Our DL models (F1-scores of 0.91, 0.90 and 0.89 for TABoLiSTM (BioBERT), TABoLiSTM (BERT) and MetaboListem, respectively) surpass by a considerable margin previous methods (using CRFs) for metabolite NER which achieved maximum F1-scores of 0.78 (8) and 0.87 (9). Our methods, compared to similar methods for chemical entity recognition (17,18,24) and metabolite NER (8,9), have the benefit of having been trained not only on abstracts and titles, but also on full-text paragraphs that are relevant for information retrieval from studies (results and discussion sections). The corpora these algorithms were trained and evaluated on are made available for future re-use and algorithm development, and to the best of our knowledge are the first of its kind for metabolomics.…”

Section: Discussionmentioning

confidence: 84%

“…The corpora these algorithms were trained and evaluated on are made available for future re-use and algorithm development, and to the best of our knowledge are the first of its kind for metabolomics. In addition, we developed a semiautomated annotation pipeline combining regular expression and rule-based approaches with dictionary searching, both for processing the training data for and as an alternative to DL methods, which too achieved outstanding performance (F1-score >0.88) which also surpassed prior DL algorithms (8, 9).…”

Section: Discussionmentioning

confidence: 99%

“…Such efficient and automatic knowledge mining systems can provide a comprehensive viewpoint to researchers by analysing a vast amount of articles. Some advances in metabolomics have been made for using computer algorithms to read scientific text (8)(9)(10)(11), however these suffer from a few drawbacks such as using rulebased annotation methods and dictionary matching. While these methods can indeed correctly capture a vast amount of metabolite names, they are dependent on the library of compound names used -and no library is exhaustive.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition

Yeung

Beck

Posma

2022

Preprint

View full text Add to dashboard Cite

Reviewing the metabolomics literature is becoming increasingly difficult because of the rapid expansion of relevant journal literature. Text-mining technologies are therefore needed to facilitate more efficient literature review. Here we contribute a standardised corpus of full-text publications from metabolomics studies and describe the development of two new metabolite named entity recognition (NER) methods. We introduce two deep learning methods for metabolite NER based on Bidirectional Long Short-Term Memory (BiLSTM) networks incorporating different transfer learning techniques. Our first model (MetaboListem) follows prior methodology using GloVe word embeddings. Our second model exploits BERT and BioBERT for embedding and is named TABoLiSTM (Transformer-Affixed BiLSTM). The methods are trained on a novel corpus annotated using rule-based methods, and evaluated on manually annotated metabolomics articles. MetaboListem (F1 score 0.890, precision 0.892, recall 0.888) and TABoLiSTM (BioBERT version: F1 score 0.909, precision 0.926, recall 0.893) have achieved state-of-the-art performance on metabolite NER. A corpus with >1,200 full-text Open Access metabolomics publications and >116,000 annotated metabolites was created. This work demonstrates that deep learning algorithms are capable of identifying metabolite names accurately and efficiently in text. The proposed corpus and NER algorithms can be used for metabolomics text-mining tasks such as information retrieval, document classification and literature-based discovery.AvailabilityThe corpus and NER algorithms are freely available with detailed instructions from Github at https://github.com/omicsNLP/MetaboliteNER.

show abstract

Section: Discussionmentioning

confidence: 84%

Section: Discussionmentioning

confidence: 84%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition

Yeung

Beck

Posma

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The resulting metabolite NER tool consisted of a hybrid strategy using dictionary-lookup and CRF-based tagger that, according to their evaluation, was able to recognize metabolite mentions with a balanced F-measure of 78.49% and precision of 83.02%. Similarly, another hybrid metabolite NER system combining mainly dictionary-lookup with CRFs was recently presented by Kongburan et al…”

Section: Integration Of Chemical and Biological Datamentioning

confidence: 99%

Information Retrieval and Text Mining Technologies for Chemistry

et al. 2017

View full text Add to dashboard Cite

Efficient access to chemical information contained in scientific literature, patents, technical reports, or the web is a pressing need shared by researchers and patent attorneys from different chemical disciplines. Retrieval of important chemical information in most cases starts with finding relevant documents for a particular chemical compound or family. Targeted retrieval of chemical documents is closely connected to the automatic recognition of chemical entities in the text, which commonly involves the extraction of the entire list of chemicals mentioned in a document, including any associated information. In this Review, we provide a comprehensive and in-depth description of fundamental concepts, technical implementations, and current technologies for meeting these information demands. A strong focus is placed on community challenges addressing systems performance, more particularly CHEMDNER and CHEMDNER patents tasks of BioCreative IV and V, respectively. Considering the growing interest in the construction of automatically annotated chemical knowledge bases that integrate chemical information and biological data, cheminformatics approaches for mapping the extracted chemical names into chemical structures and their subsequent annotation together with text mining applications for linking chemistry with biological information are also presented. Finally, future trends and current challenges are highlighted as a roadmap proposal for research in this emerging field.

show abstract

MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition

2022

View full text Add to dashboard Cite

Reviewing the metabolomics literature is becoming increasingly difficult because of the rapid expansion of relevant journal literature. Text-mining technologies are therefore needed to facilitate more efficient literature reviews. Here we contribute a standardised corpus of full-text publications from metabolomics studies and describe the development of two metabolite named entity recognition (NER) methods. These methods are based on Bidirectional Long Short-Term Memory (BiLSTM) networks and each incorporate different transfer learning techniques (for tokenisation and word embedding). Our first model (MetaboListem) follows prior methodology using GloVe word embeddings. Our second model exploits BERT and BioBERT for embedding and is named TABoLiSTM (Transformer-Affixed BiLSTM). The methods are trained on a novel corpus annotated using rule-based methods, and evaluated on manually annotated metabolomics articles. MetaboListem (F1-score 0.890, precision 0.892, recall 0.888) and TABoLiSTM (BioBERT version: F1-score 0.909, precision 0.926, recall 0.893) have achieved state-of-the-art performance on metabolite NER. A training corpus with full-text sentences from >1000 full-text Open Access metabolomics publications with 105,335 annotated metabolites was created, as well as a manually annotated test corpus (19,138 annotations). This work demonstrates that deep learning algorithms are capable of identifying metabolite names accurately and efficiently in text. The proposed corpus and NER algorithms can be used for metabolomics text-mining tasks such as information retrieval, document classification and literature-based discovery and are available from the omicsNLP GitHub repository.

show abstract

Metabolite Named Entity Recognition: A Hybrid Approach

Cited by 3 publications

References 11 publications

MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition

MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition

Information Retrieval and Text Mining Technologies for Chemistry

MetaboListem and TABoLiSTM: Two Deep Learning Algorithms for Metabolite Named Entity Recognition

Contact Info

Product

Resources

About