Krallinger, Martin
Rabal, Obdulia
Miranda-Escalada, Antonio
Valencia, Alfonso
2021-06-29
<p>Gold Standard annotations of the DrugProt corpus (training and development sets)</p>
<p><br>
</p>
<p><strong>Introduction</strong></p>
<p>The aim of the DrugProt track (similar to the previous CHEMPROT task of BioCreative VI) is to promote the development and evaluation of systems that are able to automatically detect in relations between chemical compounds/drug and genes/proteins. We have therefore generated a manually annotated corpus, the <em>DrugProt corpus</em>, where domain experts have exhaustively labeled:(a) all chemical and gene mentions, and (b) all binary relationships between them corresponding to a specific set of biologically relevant relation types (<em>DrugProt relation classes</em>). There is also an increasing interested in the integration of chemical and biomedical data understood as curation of relationships between biological and chemical entities from text and storing such information in form of structured annotation databases. Such databases are of key relevance not only for biological but also for pharmacological and clinical research. A range of different types chemical-protein/gene interactions are of key relevance for biology, including metabolic relations (e.g. substrates, products) inhibition, binding or induction associations.</p>
<p>The DrugProt track aims to address these needs and to promote the development of systems able to extract chemical-protein interactions that might be of relevance for precision medicine as well as for drug discovery and basic biomedical research.</p>
<p>The DrugProt track in BioCreative VII (BC VII) will explore recognition of chemical-protein entity relations from abstracts.</p>
<p>Teams participating in this track are provided with:</p>
<ul>
<li>PubMed abstracts</li>
<li>Manually annotated chemical compound mentions</li>
<li>Manually annotated gene/protein mentions</li>
<li>Manually annotated chemical compound-protein relations</li>
</ul>
<p> </p>
<p><strong>Zip structure:</strong></p>
<ul>
<li>Training set folder with
<ul>
<li>drugprot_training_abstracts.tsv: PubMed records</li>
<li>drugprot_training_entities.tsv: manually labeled mention annotations of chemical compounds and genes/proteins</li>
<li>drugprot_training_relations.tsv: chemical-­protein relation annotations</li>
</ul>
</li>
<li>Development set folder with
<ul>
<li>drugprot_development_abstracts.tsv</li>
<li>drugprot_development_entities.tsv</li>
<li>drugprot_development_relations.tsv</li>
</ul>
</li>
</ul>
<p> </p>
<p><strong>Data format description</strong></p>
<p>The <strong>input text files</strong> for the DrugProt track will be plain-text, UTF8-encoded PubMed records in a tab-separated format with the following three columns:</p>
<ol>
<li>Article identifier (PMID, PubMed identifier)</li>
<li>Title of the article</li>
<li>Abstract of the article</li>
</ol>
<p> </p>
<p>DrugProt <strong>entity mention annotation files</strong> contain manually labeled mention annotations of chemical compounds and genes/proteins. Such files consist of tab-separated fields containing the following six columns:</p>
<ol>
<li>Article identifier (PMID)</li>
<li>Term number (for this record)</li>
<li>Type of entity mention (CHEMICAL, GENE-Y, GENE-N)</li>
<li>Start character offset of the entity mention</li>
<li>End character offset of the entity mention</li>
<li>Text string of the entity mention</li>
</ol>
<p>Each line contains one entity, and <em>each entity is uniquely identified by its PMID and the Term Number</em>. Besides, each annotation contains an annotation type, the start-offset -the index of the first character of the annotated span in the text-, the end-offset -the index of the first character after the annotated span- and the text spanned by the annotation.</p>
<p>Example DrugProt <em>training</em> entity mention annotations:</p>
<pre><code>11808879 T1 GENE-Y 1860 1866 KIR6.2
11808879 T2 GENE-N 1993 2016 glutamate dehydrogenase
11808879 T3 GENE-Y 2242 2253 glucokinase
23017395 T1 CHEMICAL 216 223 HMG-CoA
23017395 T2 CHEMICAL 258 261 EPA</code></pre>
<p> </p>
<p>Example DrugProt <em>development</em> entity mention annotations (no distinction between GENE-Y and GENE-N):</p>
<pre><code>11808879 T1 GENE 1860 1866 KIR6.2
11808879 T2 GENE 1993 2016 glutamate dehydrogenase
11808879 T3 GENE 2242 2253 glucokinase
23017395 T1 CHEMICAL 216 223 HMG-CoA
23017395 T2 CHEMICAL 258 261 EPA</code></pre>
<p><br>
DrugProt <strong>relation annotations</strong> will be distributed as a file that contains the detailed chemical-protein relation annotations prepared for the DrugProt track. It consists of tab-separated columns containing:</p>
<ol>
<li>Article identifier (PMID)</li>
<li>DrugProt relation</li>
<li>Interactor argument 1 (<em>of type CHEMICAL</em>)</li>
<li>Interactor argument 2 (<em>of type GENE</em>)</li>
</ol>
<p>Each line contains one relation, and <em>each relation is identified by the PMID, the relation type and the two related entities</em>. In the below example, to find the entities involved in the first relation, you must find the entities with Term Identifier T1 and T52 <em>within the PMID 12488248.</em></p>
<p>Example DrugProt relation annotations:</p>
<pre><code>12488248 INHIBITOR Arg1:T1 Arg2:T52
12488248 INHIBITOR Arg1:T2 Arg2:T52
23220562 ACTIVATOR Arg1:T12 Arg2:T42
23220562 ACTIVATOR Arg1:T12 Arg2:T43
23220562 INDIRECT-DOWNREGULATOR Arg1:T1 Arg2:T14</code></pre>
<p> </p>
<p>Please, cite:</p>
<p>@inproceedings{krallinger2017overview, title={Overview of the BioCreative VI chemical-protein interaction Track}, author={Krallinger, Martin and Rabal, Obdulia and Akhondi, Saber A and P{\'e}rez, Mart{\i}n P{\'e}rez and Santamar{\'\i}a, Jes{\'u}s and Rodr{\'\i}guez, Gael P{\'e}rez and others}, booktitle={Proceedings of the sixth BioCreative challenge evaluation workshop}, volume={1}, pages={141--146}, year={2017}}</p>
<p> </p>
<p><strong>Summary statistics:</strong></p>
<pre><code> Training set Development set
Documents 3500 750
Tokens 1001168 199620
Annotated Entities 89529 18858
Annotated Relations 17288 3765</code></pre>
<p> </p>
<p>Annotated Entities:</p>
<pre><code class="language-html"> Training Entities Development Entities
CHEMICAL 46274 9853
GENE-Y [Normalizable] 28421 -
GENE-N [Non-Normalizable] 14834 -
Gene Total (N+Y) 43255 9005
Total 89529 18858</code></pre>
<p> </p>
<p>Annotated Relations:</p>
<pre><code> Training Relations Development Relations
INDIRECT-DOWNREGULATOR 1330 332
INDIRECT-UPREGULATOR 1379 302
DIRECT-REGULATOR 2250 458
ACTIVATOR 1429 246
INHIBITOR 5392 1152
AGONIST 659 131
AGONIST-ACTIVATOR 29 10
AGONIST-INHIBITOR 13 2
ANTAGONIST 972 218
PRODUCT-OF 921 158
SUBSTRATE 2003 495
SUBSTRATE_PRODUCT-OF 25 3
PART-OF 886 258
Total 17288 3765</code></pre>
<p> </p>
<p>For further information, please visit <a href="https://biocreative.bioinformatics.udel.edu/tasks/biocreative-vii/track-1/">https://biocreative.bioinformatics.udel.edu/tasks/biocreative-vii/track-1/</a> or email us at krallinger.martin@gmail.com and antoniomiresc@gmail.com</p>
<p> </p>
<p><strong>Related resources:</strong></p>
<ul>
<li><a href="https://biocreative.bioinformatics.udel.edu/tasks/biocreative-vii/track-1/">Web</a></li>
<li><a href="https://github.com/tonifuc3m/drugprot-evaluation-library">Evaluation library</a></li>
<li><a href="https://doi.org/10.5281/zenodo.4957137">Relation annotation guidelines</a></li>
<li><a href="https://doi.org/10.5281/zenodo.4957576">Gene and protein annotation guidelines</a></li>
<li><a href="https://doi.org/10.5281/zenodo.4957518">Chemicals and drugs annotation guidelines</a></li>
<li><a href="https://doi.org/10.5281/zenodo.5042178">FAQ</a></li>
</ul>
DrugProt corpus is promoted by the Plan de Impulso de las Tecnologías del Lenguaje de la Agenda Digital (Plan TL).
https://doi.org/10.5281/zenodo.5042151
oai:zenodo.org:5042151
eng
Zenodo
https://zenodo.org/communities/medicalnlp
https://doi.org/10.5281/zenodo.4955410
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
NLP
relation extraction
NER
biomedical NLP
biocreative
DrugProt corpus: Biocreative VII Track 1 - Text mining drug and chemical-protein interactions
info:eu-repo/semantics/other