Datasets:
language: en
license: mit
size_categories:
- 1K<n<10K
task_categories:
- token-classification
task_ids:
- named-entity-recognition
pretty_name: CheMBL-CL
dataset_info:
features:
- name: PMCID
dtype: string
- name: sentence
dtype: string
- name: entities
list:
- name: end
dtype: int64
- name: label
dtype: string
- name: start
dtype: int64
- name: text
dtype: string
splits:
- name: train
num_bytes: 1783667
num_examples: 8622
download_size: 733872
dataset_size: 1783667
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
tags:
- cell-line
- biomedical
- ner
ChEMBL CellLine NER Data
Dataset Description
This dataset has been extracted from Europe PMC (EPMC), a free database offering comprehensive access to life sciences research literature. EPMC aggregates content from various sources, including PubMed, arXiv, and other repositories, and provides open access to millions of scientific articles. This dataset has been generated as part of a project collaboration between Europe PMC, Open Targets, and ChEMBL] at EMBL-EBI. Standard: Silver
Dataset Details
The dataset was extracted using assay descriptions from CheMBL andThe dataset has been annotated on mentions of cell lines. This makes it a valuable resource as it can be used for(but not limited to) the following downstream natural language processing (NLP) tasks in the biomedical domain.
NLP Tasks
Named Entity Recognition (NER): Identify and classify mentions of cell lines or related entities appearing in biomedical contexts.
Relationship Extraction: Extract relationships between cell lines and other biomedical entities, such as genes, diseases, or drugs.
Text Classification: Classify sentences or articles based on their relevance to specific cell lines, particularly in cancer research or drug development.
Sentiment Analysis: Analyze the sentiment or tone of texts discussing cell lines, such as the evaluation of experimental results (positive or negative).
Information Retrieval: Develop systems to retrieve articles or specific mentions of cell lines based on user queries.
Entity Linking: Link cell line mentions in text to standardized identifiers in cell line ontologies or databases.
Question Answering (QA): Build systems that can answer specific questions about cell lines, such as their role in particular diseases or experiments.
Topic Modeling: Analyze the dataset to uncover major themes or trends in research involving cell lines.
Text Summarization: Automatically generate summaries of articles or sections discussing cell lines.
Who funded the creation of the dataset?
Any other comments?
The dataset aims to provide a foundational resource for advancing NLP in biomedicine.
Dataset Composition
- What experiments were initially run on this dataset? No experiments have been conducted yet. Updates will follow as they occur.
Data Collection Process
How was the data collected? The dataset was collected using the Europe PMC API. Articles marked as "open access" were retrieved, and those labeled as "retraction of publication" were excluded. Duplicate entries were filtered by ensuring unique PMCIDs.
Who was involved in the data collection process? The data collection process was carried out by researchers at EMBL-EBI, leveraging automated tools for querying and processing the Europe PMC repository.
Are there any known errors, sources of noise, or redundancies in the data? None have been identified yet.
Data Preprocessing
What preprocessing/cleaning was done?
- Only open-access articles were retrieved.
- Articles labeled as "retraction of publication" were excluded.
- Duplicate entries based on the PMCID column were removed.
- Paragraph text from each section of an article was Extracted with the relevant section referenced in the 'Section Column'
- Extra whitespace, inlne math/latex formatting and irrelevant sections such "Disclosure", "Publisher's note", etc, were filtered
- Name identifiers and personal data was also removed from the dataset
Dataset Distribution
How is the dataset distributed? The dataset is freely available for use and reproduction. Proper citation of the authors is required(Information to be updated).
When will the dataset be released/first distributed? To be updated.
Dataset Maintenance
Who is supporting/hosting/maintaining the dataset?
Europe PMC, and ChEMBL team responsible for the dataset's maintenance.
How does one contact the owner/curator/manager of the dataset?
Contact can be made via the community discussion forums on GitHub or Hugging Face.
Will the dataset be updated?
Yes, updates will occur as the project progresses.
How often and by whom?
Updates will be carried out periodically by the team.
How will updates/revisions be documented and communicated?
Updates will be documented via GitHub, using version tags.
Is there a repository to link to any/all papers/systems that use this dataset?
Yes, a GitHub repository will track publications and systems using this dataset.
If others want to extend/augment/build on this dataset, is there a mechanism for them to do so?
Yes, contributions are encouraged via GitHub. Quality will be assessed through pull requests, and accepted contributions will be communicated to users via version tags and release notes.