HOME- R&D
- Bryn Mawr Conference
- Workshops & Training
- Program
- Drug Discovery Innovation
- Abstracts
- Speakers
- Blank, L
- Boyer, S
- Cook, J
- Gibbs, A
- Hopfgartner, G
- Lett, G.S.
- Lobanov, V
- Mosenkis, D
- Nieuwenhuis, J
- Wikel, J
- Wiseman, J
- Zauhar, R
- Proteins
- Quantum Biochemistry
- Screening
- Web Services
- Pharmacophores
- Graph Mining
- Nanotech
- Membranes & Ion Channels
- Exhibition
- Registration
- Jobs
- Contact
- Schedule
|
|
|
|
|
|
|
|
Analysis based on Molecular Diversity
Steve Boyer, IBM
We have successfully demonstrated the use of IBM's text analytics & annotation technology to electronically analyze the full-text of > 2,602,697 patent documents and created a database of millions of chemical structures. In doing so, we have opened the doors to searching and documenting molecular entities in a previously inaccessible area characterized by arcane and inconsistent nomenclature. By converting previously obscure chemical nomenclatures into universally recognized structures, the vast scientific literature of chemical and molecular information is rendered amenable to database technologies.
This accomplishment was demonstrated by using IBM's text analytic technology to electronically 'read' over 25 years of patent full-text documents, and selectively identify the chemical entities in these documents. Once identified, the chemical names were extracted and post-processed to convert the chemical names into chemical structures (SMILES strings and /or mol files). This step effectively renders the scientific literature searchable by structure/substructure search applications. The combined technologies for reading and processing molecular structure will allow researchers the ability to build large databases of previously inaccessible literature - relevant in the areas of patents, pharmaceuticals, publishing, health care, and environmental science, to cite just a few examples. Current work on the molecular analysis of the patent corpus will be discussed.
|
|
|
|
|
|
|
|
|
