phucdev commited on
Commit
f4e2664
1 Parent(s): 2374f62

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +165 -57
README.md CHANGED
@@ -1,57 +1,165 @@
1
- ---
2
- dataset_info:
3
- features:
4
- - name: NofPmids
5
- dtype: float64
6
- - name: NofSnps
7
- dtype: float64
8
- - name: associationType
9
- dtype: string
10
- - name: diseaseId
11
- dtype: string
12
- - name: diseaseName
13
- dtype: string
14
- - name: diseaseType
15
- dtype: string
16
- - name: disease_mention
17
- dtype: string
18
- - name: geneId
19
- dtype: string
20
- - name: geneSymbol
21
- dtype: string
22
- - name: gene_mention
23
- dtype: string
24
- - name: originalSource
25
- dtype: string
26
- - name: pmid
27
- dtype: int64
28
- - name: raw_sentence
29
- dtype: string
30
- - name: score
31
- dtype: float64
32
- - name: sentence
33
- dtype: string
34
- - name: source
35
- dtype: string
36
- splits:
37
- - name: train
38
- num_bytes: 1907978
39
- num_examples: 4000
40
- - name: validation
41
- num_bytes: 1134075
42
- num_examples: 2400
43
- - name: test
44
- num_bytes: 756401
45
- num_examples: 1600
46
- download_size: 2013053
47
- dataset_size: 3798454
48
- configs:
49
- - config_name: default
50
- data_files:
51
- - split: train
52
- path: data/train-*
53
- - split: validation
54
- path: data/validation-*
55
- - split: test
56
- path: data/test-*
57
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ dataset_info:
3
+ features:
4
+ - name: NofPmids
5
+ dtype: float64
6
+ - name: NofSnps
7
+ dtype: float64
8
+ - name: associationType
9
+ dtype: string
10
+ - name: diseaseId
11
+ dtype: string
12
+ - name: diseaseName
13
+ dtype: string
14
+ - name: diseaseType
15
+ dtype: string
16
+ - name: disease_mention
17
+ dtype: string
18
+ - name: geneId
19
+ dtype: string
20
+ - name: geneSymbol
21
+ dtype: string
22
+ - name: gene_mention
23
+ dtype: string
24
+ - name: originalSource
25
+ dtype: string
26
+ - name: pmid
27
+ dtype: int64
28
+ - name: raw_sentence
29
+ dtype: string
30
+ - name: score
31
+ dtype: float64
32
+ - name: sentence
33
+ dtype: string
34
+ - name: source
35
+ dtype: string
36
+ splits:
37
+ - name: train
38
+ num_bytes: 1907978
39
+ num_examples: 4000
40
+ - name: validation
41
+ num_bytes: 1134075
42
+ num_examples: 2400
43
+ - name: test
44
+ num_bytes: 756401
45
+ num_examples: 1600
46
+ download_size: 2013053
47
+ dataset_size: 3798454
48
+ configs:
49
+ - config_name: default
50
+ data_files:
51
+ - split: train
52
+ path: data/train-*
53
+ - split: validation
54
+ path: data/validation-*
55
+ - split: test
56
+ path: data/test-*
57
+ task_categories:
58
+ - text-classification
59
+ language:
60
+ - en
61
+ tags:
62
+ - medical
63
+ - biology
64
+ - relation-classification
65
+ - relation-extraction
66
+ - gene
67
+ - disease
68
+ pretty_name: GDA
69
+ size_categories:
70
+ - 1K<n<10K
71
+ ---
72
+ # Dataset Card for GDA
73
+
74
+ ## Dataset Description
75
+
76
+ - **Repository:** https://github.com/EsmaeilNourani/Deep-GDAE
77
+ - **Paper:** [Association extraction from biomedical literature based on representation and transfer learning](https://www.sciencedirect.com/science/article/abs/pii/S0022519319304813)
78
+
79
+ #### Dataset Summary
80
+
81
+ <!-- Provide a quick summary of the dataset. -->
82
+ **GDA Dataset Summary:**
83
+
84
+ Nourani and Reshadata (2020) developed a dataset called GDA corpus as a sentence-level evaluation dataset for extracting the association between genes and diseases based on some efficient databases.
85
+ They used DisGeNET, a database of Gene-Disease-Associations (GDAs), and PubTator to retrieve biomedical texts (PubMed abstracts).
86
+ Using PubTator, they found all the PMIDs containing at least one gene and disease name.
87
+ Samples of the true class were extracted from DisGeNET, considering only curated associations.
88
+ For the creation of non-associated or false samples, a systematic three-step filtering process was used to ensure high-quality annotations.
89
+ This included the exclusion of known associations from DisGeNET and CTD, as well as a linguistic filtering step to remove sentences that linguistically imply an association.
90
+ GDA was constructed automatically and contains 8000 sentences with 1904 and 3635 unique diseases and genes respectively.
91
+
92
+
93
+ ### Languages
94
+ The language in the dataset is English.
95
+
96
+ ## Dataset Structure
97
+ <!-- This section provides a description of the dataset fields, and additional information about the dataset structure such as criteria used to create the splits, relationships between data points, etc. -->
98
+ ### Dataset Instances
99
+ An example of 'train' looks as follows:
100
+ ```json
101
+ {
102
+ "NofPmids": 2.0,
103
+ "NofSnps": 0.0,
104
+ "associationType": "Biomarker",
105
+ "diseaseId": "C0043459",
106
+ "diseaseName": "Zellweger Syndrome",
107
+ "diseaseType": "disease",
108
+ "disease_mention": "H-02",
109
+ "geneId": "5194",
110
+ "geneSymbol": "PEX13",
111
+ "gene_mention": "PEX13",
112
+ "originalSource": "CTD_human",
113
+ "pmid": 10332040,
114
+ "raw_sentence": "We now have evidence that the complete human cDNA encoding Pex13p, an SH3 protein of a docking factor for the peroxisome targeting signal 1 receptor (Pex5p), rescues peroxisomal matrix protein import and its assembly in fibroblasts from PBD patients of complementation group H. In addition, we detected mutations on the human PEX13 cDNA in two patients of group H. A severe phenotype of a ZS patient (H-02) was homozygous for a nonsense mutation, W234ter, which results in the loss of not only the SH3 domain but also the putative transmembrane domain of Pex13p.",
115
+ "score": 0.400549453568426,
116
+ "sentence": "We now have evidence that the complete human cDNA encoding Pex13p, an SH3 protein of a docking factor for the peroxisome targeting signal 1 receptor (Pex5p), rescues peroxisomal matrix protein import and its assembly in fibroblasts from PBD patients of complementation group H. In addition, we detected mutations on the human <span class=\"gene\" id=\"10332040-3-326-331\">PEX13</span> cDNA in two patients of group H. A severe phenotype of a <span class=\"disease\" id=\"10332040-3-389-391\">ZS</span> patient (<span class=\"disease\" id=\"10332040-3-401-405\">H-02</span>) was homozygous for a nonsense mutation, W234ter, which results in the loss of not only the SH3 domain but also the putative transmembrane domain of Pex13p.",
117
+ "source": "CTD_human;ORPHANET"
118
+ }
119
+ ```
120
+
121
+ ### Data Fields
122
+
123
+ Here's the Data Fields section for the GDA corpus based on the dataset features provided:
124
+
125
+ - `NofPmids`: the number of PubMed IDs related to the gene-disease association, stored as a `float64` feature.
126
+ - `NofSnps`: the number of single nucleotide polymorphisms (SNPs) related to the gene-disease association, stored as a `float64` feature.
127
+ - `associationType`: the type of association (e.g., Negative, Biomarker, Therapeutic) between the gene and the disease, a `string` feature.
128
+ - `diseaseId`: the unique identifier for the disease discussed, a `string` feature.
129
+ - `diseaseName`: the name of the disease, a `string` feature.
130
+ - `diseaseType`: the type of the disease (e.g., disease, group, phenotype), a `string` feature.
131
+ - `disease_mention`: the specific mention of the disease within the source text, a `string` feature.
132
+ - `geneId`: the unique identifier for the gene discussed, a `string` feature.
133
+ - `geneSymbol`: the symbol representing the gene, a `string` feature.
134
+ - `gene_mention`: the specific mention of the gene within the source text, a `string` feature.
135
+ - `originalSource`: the original source, a `string` feature.
136
+ - `pmid`: the PubMed ID associated with the citation from which the sentence is derived, an `int64` feature.
137
+ - `raw_sentence`: the original sentence from the source document, a `string` feature.
138
+ - `score`: a score reflecting the confidence or relevance of the association between the gene and the disease, stored as a `float64` feature.
139
+ - `sentence`: the sentence with span annotation, a `string` feature.
140
+ - `source`: the database or repository from which the association data was taken, a `string` feature.
141
+
142
+ ## Citation
143
+ <!-- If there is a paper or blog post introducing the dataset, the APA and Bibtex information for that should go in this section. -->
144
+
145
+ **BibTeX:**
146
+ ```
147
+ @article{NOURANI2020110112,
148
+ title = {Association extraction from biomedical literature based on representation and transfer learning},
149
+ journal = {Journal of Theoretical Biology},
150
+ volume = {488},
151
+ pages = {110112},
152
+ year = {2020},
153
+ issn = {0022-5193},
154
+ doi = {https://doi.org/10.1016/j.jtbi.2019.110112},
155
+ url = {https://www.sciencedirect.com/science/article/pii/S0022519319304813},
156
+ author = {Esmaeil Nourani and Vahideh Reshadat},
157
+ keywords = {Gene-Disease Association Extraction, Attention Mechanism, BioBERT}
158
+ }
159
+ ```
160
+
161
+ **APA:**
162
+ - Nourani, E., & Reshadat, V. (2020). Association extraction from biomedical literature based on representation and transfer learning. Journal of Theoretical Biology, 488, 110112. https://doi.org/10.1016/j.jtbi.2019.110112
163
+
164
+ ## Dataset Card Authors
165
+ [@phucdev](https://github.com/phucdev)