Dataset Information

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 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.

SUBMITTER: Lai PT

PROVIDER: S-EPMC6755615 | biostudies-literature | 2018 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Statistical principle-based approach for gene and protein related object recognition.

Lai Po-Ting PT Huang Ming-Siang MS Yang Ting-Hao TH Hsu Wen-Lian WL Tsai Richard Tzong-Han RT

Journal of cheminformatics 20181217 1

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 ...[more]

PMID: 30560325

Similar Datasets

Project description:BackgroundObject detection is a new artificial intelligence approach to morphological recognition and labeling parasitic pathogens. Due to the lack of equipment and trained personnel, artificial intelligence innovation for searching various parasitic products in stool examination will enable patients in remote areas of undeveloped countries to access diagnostic services. Because object detection is a developing approach that has been tested for its effectiveness in detecting intestinal parasitic objects such as protozoan cysts and helminthic eggs, it is suitable for use in rural areas where many factors supporting laboratory testing are still lacking. Based on the literatures, the YOLOv4-Tiny produces faster results and uses less memory with the support of low-end GPU devices. In comparison to the YOLOv3 and YOLOv3-Tiny models, this study aimed to propose an automated object detection approach, specifically the YOLOv4-Tiny model, for automatic recognition of intestinal parasitic products in stools.MethodsTo identify protozoan cysts and helminthic eggs in human feces, the three YOLO approaches; YOLOv4-Tiny, YOLOv3, and YOLOv3-Tiny, were trained to recognize 34 intestinal parasitic classes using training of image dataset. Feces were processed using a modified direct smear method adapted from the simple direct smear and the modified Kato-Katz methods. The image dataset was collected from intestinal parasitic objects discovered during stool examination and the three YOLO models were trained to recognize the image datasets.ResultsThe non-maximum suppression technique and the threshold level were used to analyze the test dataset, yielding results of 96.25% precision and 95.08% sensitivity for YOLOv4-Tiny. Additionally, the YOLOv4-Tiny model had the best AUPRC performance of the three YOLO models, with a score of 0.963.ConclusionThis study, to our knowledge, was the first to detect protozoan cysts and helminthic eggs in the 34 classes of intestinal parasitic objects in human stools.

Project description:The recognition of novel objects is a common cognitive test for rodents, but current paradigms have limitations, such as low sensitivity, possible odor confounds and stress due to being performed outside of the homecage. We have developed a paradigm that takes place in the homecage and utilizes four stimuli per trial, to increase sensitivity. Odor confounds are eliminated because stimuli consist of inexpensive, machined wooden beads purchased in bulk, so each experimental animal has its own set of stimuli. This paradigm consists of three steps. In Step 1, the sampling phase, animals freely explore familiar objects (FO). Novel Objects (NO1 and NO2) are soiled with bedding from the homecage, to acquire odor cues identical to those of the FO. Steps 2 and 3 are test phases. Herein we report results of this paradigm from neurologically intact adult rats and mice of both sexes. Identical procedures were used for both species, except that the stimuli used for the mice were smaller. As expected in Step 2 (NO1 test phase), male and female rats and mice explored NO1 significantly more than FO. In Step 3 (NO2 test phase), rats of both sexes demonstrated a preference for NO2, while this was seen only in female mice. These results indicate robust novelty recognition during Steps 2 and 3 in rats. In mice, this was reliably seen only in Step 2, indicating that Step 3 was difficult for them under the given parameters. This paradigm provides flexibility in that length of the sampling phase, and the delay between test and sampling phases can be adjusted, to tailor task difficulty to the model being tested. In sum, this novel object recognition test is simple to perform, requires no expensive supplies or equipment, is conducted in the homecage (reducing stress), eliminates odor confounds, utilizes 4 stimuli to increase sensitivity, can be performed in both rats and mice, and is highly flexible, as sampling phase and the delay between steps can be adjusted to tailor task difficulty. Collectively, these results indicate that this paradigm can be used to quantify novel object recognition across sex and species.

Dataset Information

Statistical principle-based approach for gene and protein related object recognition.

Publications

Statistical principle-based approach for gene and protein related object recognition.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets