Presentation is loading. Please wait.

Presentation is loading. Please wait.

Next Steps in Literature Mining Marti Hearst UC Berkeley ASIST 2003 Literature Mining Panel.

Similar presentations


Presentation on theme: "Next Steps in Literature Mining Marti Hearst UC Berkeley ASIST 2003 Literature Mining Panel."— Presentation transcript:

1 Next Steps in Literature Mining Marti Hearst UC Berkeley ASIST 2003 Literature Mining Panel

2 Literature Mining Goals Discover new information … … As opposed to discovering which statistical patterns characterize occurrence of known information. Method: Use large text collections to gather evidence to support (or refute) hypotheses Extract facts Make connections Draw inferences

3 Outline Don’t Repeat History Use Time Machines More Ambitious Semantics

4 Don’t Repeat History Don’t show the obvious e.g., Cheney is president Don’t show what you’ve already shown Only show the most recent version of information Show which information is not present Changes in the usual pattern Something stops happening

5 Create “Time Machines” Do systematic analyses of how to find out what is now known based on what used to be known. Example: See if information that was found via microarray analysis could have been found in the literature before its invention. Reverse-engineer the method and use it to find new information. Different approach in a new paper by Lukose, Adar, and Chan of HP Labs

6 More Ambitious Semantics Go beyond extracting entities Find relations between entities Convert clauses into propositions

7 : Mouse Bim proteins (isoforms EL, L, S) binds to human Bcl-2 (bacteriophoage screening using cDNA expression library from T-Lymphoma cell line KO52DA20). Human BimEL protein is 89% identical to mouse BimEL, Human BimL is 85% identical to mouse BimL (Hybridization of mouse bim cDNA to human fetal spleen and peripheral blood cDNA library). Bim mRNA is detected in B and T lyphoid cells (Northern blot analysis of mouse KO52DA20, WEHI 703, WEHI 707, WEHI7.1, CH1, WEHI231 WEHI415, B6.23.16BW2 cell extracts). BimL protein interact with Bcl-2 OR Bcl-XL, or Bcl-w proteins (Immuno- precipitation (anti-Bcl-2 OR Bcl-XL OR Bcl-w)) followed by Western blot (anti- EEtag) using extracts human 293T cells co-transfected with EE-tagged BimL AND (bcl-2 OR bcl-XL OR bcl-w) plasmids) BimL deleted of the BH3 domain does not bind to Bcl-2 OR Bcl-XL, or Bcl-w proteins (under experimental conditions mentioned above)

8 Death Receptors Signaling Survival Factors Signaling Ca ++ Signaling P53 pathway Caspase 12 Effecter Caspases (3,6,7) Caspase 9 Apaf 1 IAPs NFkB Mitochondria Cytochrome c Bax, Bak Apoptosis Bcl-2 like BH3 only Apoptosis Network Smac ER Stress Genotoxic Stress Initiator Caspases (8, 10) AIF Lost of Attachment Cell Cycle stress, etc Slide courtesy TingTing Zhang

9 More Ambitious Semantics Go beyond extracting entities Find relations between entities Convert clauses into propositions Go beyond combining via co-occurrence Draw inferences between the facts and relations Incorporate domain knowledge

10 Our Approach Assign Semantics using Statistics Hierarchical Lexical Ontologies to generalize Redundancy in the data Build up Layers of Representation Syntactic and Semantic Use these in a feedback loop

11 Goal: Convert Text to Generalized Semantics HDAC inhibitors induce differentiation of cultured murine erythroleukemia cells. [ ] [ [ ]]

12 Thank You! More information: http://biotext.berkeley.edu


Download ppt "Next Steps in Literature Mining Marti Hearst UC Berkeley ASIST 2003 Literature Mining Panel."

Similar presentations


Ads by Google