A machine learning framework for extracting information from biological pathway images in the literature

<p><strong>466 target chemicals_selected chemicals:</strong></p> <p>Target chemicals satisfying the criteria for biochemical reactions not covered by MetaNetX and KEGG.</p> <p>&nbsp;</p> <p><strong>466 target chemicals_statistics:</strong><br>Numbers of MetaCyc reactions, papers, and pathway images collected for 466 target chemicals from the bio-based chemicals map (Jang et al., Trends in Biotechnology, 2023).</p> <p>&nbsp;</p> <p><strong>arrow detection_bounding box labels:</strong><br>Bounding box labels for 6,471 images in the training and validation datasets and 100 images in the test dataset. The corresponding images are available in "arrow detection_training and validation datasets.zip".</p> <p>&nbsp;</p> <p><strong>arrow detection_test dataset:</strong></p> <p>Test dataset for arrow detection using Faster R-CNN model. A total of 100 images have been prepared from 89 papers searched through PubMed Central (PMC).</p> <p>&nbsp;</p> <p><strong>arrow detection_training and validation datasets:</strong></p> <p>Training and validation datasets for arrow detection using Faster R-CNN model. A total of 6,471 images have been prepared, including 2,332 images from five different sources and 4,139 augmented images.</p> <p>&nbsp;</p> <p><strong>EBPI outputs:</strong></p> <p>Reaction information extracted using EBPI from 49,846 biological pathway images across 466 target chemicals.</p> <p>&nbsp;</p> <p><strong>text classification_training, validation and test datasets:</strong></p> <p>Dataset for text classification using BioBERT. A total of 59,370 terms have been prepared, including 15,101 &ldquo;gene&rdquo; terms, 21,417 &ldquo;protein&rdquo; terms, and 22,852 &ldquo;others&rdquo; terms by combining the data from MetaCyc and the PaddleOCR results from the papers.</p>