Table 1
An example data structure in which the extracted biomarkers are organized for downstream research. This is a subset of the output table at https://github.com/Damonlin11/FDA-approved-Targeted-Therapies-Label-Extraction/blob/main/2021-06-27%20FDA-approved%20targeted%20therapy%20labels.csv, generated with the code at https://github.com/Damonlin11/FDA-approved-Targeted-Therapies-Label-Extraction/blob/main/FDA_therapy_biomarker_extraction.ipynb.
| geneProtein_label | therapy_label | disease_label | drug_label | drug_label |
|---|---|---|---|---|
| PD-L1 | Atezdizumab | Urothelial carcinoma | TECENTRIQ- atezolizumab injection, solution | 50242-917-01, 50242-917-86, 50242-918-01, 50242-918-86 |
| BRAF | Atezolizumab | Melanoma | TECENTRIQ- atezolizumab injection, solution | 50242-917-01, 50242-917-86, 50242-918-01, 50242-918-86 |
| PD-L1 | Atezolizumab | Non-small cell lung cancer | TECENTRIQ- atezolizumab injection, solution | 50242-917-01, 50242-917-86, 50242-918-01, 50242-918-86 |
| EGFR | Atezolizumab | Non-small cell lung cancer | TECENTRIQ- atezolizumab injection, solution | 50242-917-01, 50242-917-86, 50242-918-01, 50242-918-86 |
| ALK genomic | Atezolizumab | Non-small cell lung cancer | TECENTRIQ- atezolizumab injection, solution | 50242-917-01, 50242-917-86, 50242-918-01, 50242-918-86 |
| PD-L1 | Atezolizumab | Breast cancer | TECENTRIQ- atezdizumab injection, solution | 50242-917-01, 50242-917-86, 50242-918-01, 50242-918-86 |
| PD-L1 | Nivolumab | Non-small cell lung cancer | OPDIVO- nivolumab Injection | 0003-3734-13, 0003-3772-11, 0003-3774-12 |
| EGFR | Nivolurnab | Non-small cell lung cancer | OPDIVO- nlvolumab injection | 0003-3734-13, 0003-3772-11, 0003-3774-12 |
| ALK genomic | Nivolurnab | Non-small cell lung cancer | OPDIVO- nlvolumab injection | 0003-3734-13, 0003-3772-11, 0003-3774-12 |
Figure 1
Logical structure and output for detecting biomarker entities in URL and free text using the biomarker_nlp package.
Figure 2
Logical structure and output for detecting negation in free text using the biomarker_nlp package.