Statistical principle-based approach for gene and protein related object recognition
Abstract The large number of chemical and pharmaceutical patents has attracted researchers doing biomedical text mining to extract valuable information such as chemicals, genes and gene products. To facilitate gene and gene product annotations in patents, BioCreative V.5 organized a gene- and protei...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
BMC
2018-12-01
|
Series: | Journal of Cheminformatics |
Subjects: | |
Online Access: | http://link.springer.com/article/10.1186/s13321-018-0314-7 |
id |
doaj-ecf74773132444f29eba80067d36ac0a |
---|---|
record_format |
Article |
spelling |
doaj-ecf74773132444f29eba80067d36ac0a2020-11-25T00:41:49ZengBMCJournal of Cheminformatics1758-29462018-12-011011910.1186/s13321-018-0314-7Statistical principle-based approach for gene and protein related object recognitionPo-Ting Lai0Ming-Siang Huang1Ting-Hao Yang2Wen-Lian Hsu3Richard Tzong-Han Tsai4Department of Computer Science, National Tsing-Hua UniversityBioinformatics Program, Taiwan International Graduate Program, Institute of Information Science, Academia SinicaDepartment of Computer Science, National Tsing-Hua UniversityDepartment of Computer Science, National Tsing-Hua UniversityIntelligent Information Service Research Laboratory, Department of Computer Science and Information Engineering, National Central UniversityAbstract The large number of chemical and pharmaceutical patents has attracted researchers doing biomedical text mining to extract valuable information such as chemicals, genes and gene products. To facilitate gene and gene product annotations in patents, BioCreative V.5 organized a gene- and protein-related object (GPRO) recognition task, in which participants were assigned to identify GPRO mentions and determine whether they could be linked to their unique biological database records. In this paper, we describe the system constructed for this task. Our system is based on two different NER approaches: the statistical-principle-based approach (SPBA) and conditional random fields (CRF). Therefore, we call our system SPBA-CRF. SPBA is an interpretable machine-learning framework for gene mention recognition. The predictions of SPBA are used as features for our CRF-based GPRO recognizer. The recognizer was developed for identifying chemical mentions in patents, and we adapted it for GPRO recognition. In the BioCreative V.5 GPRO recognition task, SPBA-CRF obtained an F-score of 73.73% on the evaluation metric of GPRO type 1 and an F-score of 78.66% on the evaluation metric of combining GPRO types 1 and 2. Our results show that SPBA trained on an external NER dataset can perform reasonably well on the partial match evaluation metric. Furthermore, SPBA can significantly improve performance of the CRF-based recognizer trained on the GPRO dataset.http://link.springer.com/article/10.1186/s13321-018-0314-7Named entity recognitionInformation extractionNatural language processingBiomedical text miningMachine learningMedical chemical patent |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Po-Ting Lai Ming-Siang Huang Ting-Hao Yang Wen-Lian Hsu Richard Tzong-Han Tsai |
spellingShingle |
Po-Ting Lai Ming-Siang Huang Ting-Hao Yang Wen-Lian Hsu Richard Tzong-Han Tsai Statistical principle-based approach for gene and protein related object recognition Journal of Cheminformatics Named entity recognition Information extraction Natural language processing Biomedical text mining Machine learning Medical chemical patent |
author_facet |
Po-Ting Lai Ming-Siang Huang Ting-Hao Yang Wen-Lian Hsu Richard Tzong-Han Tsai |
author_sort |
Po-Ting Lai |
title |
Statistical principle-based approach for gene and protein related object recognition |
title_short |
Statistical principle-based approach for gene and protein related object recognition |
title_full |
Statistical principle-based approach for gene and protein related object recognition |
title_fullStr |
Statistical principle-based approach for gene and protein related object recognition |
title_full_unstemmed |
Statistical principle-based approach for gene and protein related object recognition |
title_sort |
statistical principle-based approach for gene and protein related object recognition |
publisher |
BMC |
series |
Journal of Cheminformatics |
issn |
1758-2946 |
publishDate |
2018-12-01 |
description |
Abstract The large number of chemical and pharmaceutical patents has attracted researchers doing biomedical text mining to extract valuable information such as chemicals, genes and gene products. To facilitate gene and gene product annotations in patents, BioCreative V.5 organized a gene- and protein-related object (GPRO) recognition task, in which participants were assigned to identify GPRO mentions and determine whether they could be linked to their unique biological database records. In this paper, we describe the system constructed for this task. Our system is based on two different NER approaches: the statistical-principle-based approach (SPBA) and conditional random fields (CRF). Therefore, we call our system SPBA-CRF. SPBA is an interpretable machine-learning framework for gene mention recognition. The predictions of SPBA are used as features for our CRF-based GPRO recognizer. The recognizer was developed for identifying chemical mentions in patents, and we adapted it for GPRO recognition. In the BioCreative V.5 GPRO recognition task, SPBA-CRF obtained an F-score of 73.73% on the evaluation metric of GPRO type 1 and an F-score of 78.66% on the evaluation metric of combining GPRO types 1 and 2. Our results show that SPBA trained on an external NER dataset can perform reasonably well on the partial match evaluation metric. Furthermore, SPBA can significantly improve performance of the CRF-based recognizer trained on the GPRO dataset. |
topic |
Named entity recognition Information extraction Natural language processing Biomedical text mining Machine learning Medical chemical patent |
url |
http://link.springer.com/article/10.1186/s13321-018-0314-7 |
work_keys_str_mv |
AT potinglai statisticalprinciplebasedapproachforgeneandproteinrelatedobjectrecognition AT mingsianghuang statisticalprinciplebasedapproachforgeneandproteinrelatedobjectrecognition AT tinghaoyang statisticalprinciplebasedapproachforgeneandproteinrelatedobjectrecognition AT wenlianhsu statisticalprinciplebasedapproachforgeneandproteinrelatedobjectrecognition AT richardtzonghantsai statisticalprinciplebasedapproachforgeneandproteinrelatedobjectrecognition |
_version_ |
1725285410496577536 |