StrataRx 2012

Presentation of clearScience

Transcript

1. clearScience dragging scientific communication into the information age erich s. huang, md/phd director, cancer research sage bionetworks erich.huang@sagebase.org
2. http://www.viahealthcaredesign.com/LMCH%20Universal%20Exam%20Rm.jpg
3. Betty Jones “I have a lump near my mastectomy scar”
4. Betty Jones menopause ~50 years “high risk” breast cancer “normal risk” breast cancer
5. 36 years Betty Jones menopause ~50 years “high risk” breast cancer “normal risk” breast cancer
6. 61% 39% Triple Negative Breast Cancer All Other Types 84% 16% Triple Negative Breast Cancer All Other Types Betty Jones African American Population Non-African American Population http://jama.jamanetwork.com/article.aspx?articleid=202952
7. breast cancer cell depends on receptor-driven growth signals estrogen receptor progesterone receptor her2 receptor
8. breast cancer cell depends on receptor-driven growth signals estrogen receptor progesterone receptor her2 receptor
9. breast cancer cell depends on receptor-driven growth signals estrogen receptor progesterone receptor her2 receptor GROWTH SIGNALS
10. breast cancer cell growth signals become independent of receptors
11. “triple negative” breast cancer breast cancer cell growth signals become independent of receptors
12. “triple negative” breast cancer the receptors are no longer present on the cell surface... breast cancer cell growth signals become independent of receptors
13. “triple negative” breast cancer the receptors are no longer present on the cell surface... but the intracellular signals persist breast cancer cell growth signals become independent of receptors
14. “triple negative” breast cancer the receptors are no longer present on the cell surface... but the intracellular signals persist breast cancer cell growth signals become independent of receptors INDEPENDENT GROWTH SIGNALS
15. http://www.cancer.gov/images/cdr/live/CDR415523-750.jpg
16. http://www.cancer.gov/images/cdr/live/CDR415523-750.jpg
17. http://www.cancer.gov/images/cdr/live/CDR415523-750.jpg
18. Betty Jones minus plus young lymph node negative african-american triple negative multifocal cancer
19. Betty Jones minus plus young lymph node negative african-american triple negative multifocal cancer = Aggressive therapeutic strategy
20. http://www.flickr.com/photos/kbrookes/7497906298/sizes/h/in/photostream/
21. http://radonc.ucsf.edu/patient_information/breast/images/Radiation_Oncology_breast_img_2.jpg radiotherapy 3d planning
22. Betty Jones “I have a lump near my mastectomy scar”
23. Betty Jones Triple negative breast cancer. What are the options?
24. Betty Jones Triple negative breast cancer. What are the options? Standard Stuff (Old Fashioned Targets) Anthracyclines Taxanes Anti-Metabolites
25. Betty Jones Triple negative breast cancer. What are the options? Standard Stuff (Old Fashioned Targets) Anthracyclines Taxanes Anti-Metabolites Platinum Agents
26. Betty Jones Triple negative breast cancer. What are the options? Standard Stuff (Old Fashioned Targets) Anthracyclines Taxanes Anti-Metabolites Platinum Agents { conventional chemotherapy
27. Betty Jones Triple negative breast cancer. What are the options? Standard Stuff (Old Fashioned Targets) Anthracyclines Taxanes Anti-Metabolites Platinum Agents Targeted Agents Cetuximab + Platinum (BALI-1) Bevacizumab (GeparQuinto) PARP Inhibitors SRC Inhibitors HDAC Inhibitors (ER-alpha) { conventional chemotherapy
28. Betty Jones Triple negative breast cancer. What are the options? Standard Stuff (Old Fashioned Targets) Anthracyclines Taxanes Anti-Metabolites Platinum Agents Targeted Agents Cetuximab + Platinum (BALI-1) Bevacizumab (GeparQuinto) PARP Inhibitors SRC Inhibitors HDAC Inhibitors (ER-alpha) { conventional chemotherapy { new therapies
29. Betty Jones Triple negative breast cancer. What are the options? Standard Stuff (Old Fashioned Targets) Anthracyclines Taxanes Anti-Metabolites Platinum Agents Targeted Agents Cetuximab + Platinum (BALI-1) Bevacizumab (GeparQuinto) PARP Inhibitors SRC Inhibitors HDAC Inhibitors (ER-alpha) { conventional chemotherapy { new therapies CURRENT KNOWLEDGE: ALL THESE THERAPIES ARE A WASH
30. Betty Jones
31. Betty Jones
32. http://www.viahealthcaredesign.com/LMCH%20Universal%20Exam%20Rm.jpg
33. If I have seen further it is by standing on the shoulders of giants. “ ”
36. ~4 million datapoints molecularly-customized cancer treatment plan
38. CITED 463 TIMES
39. CITED 463 TIMES 3 GENOMICS- GUIDED TRIALS
40. CITED 463 TIMES 3 GENOMICS- GUIDED TRIALS
43. ~4 million datapoints molecularly-customized cancer treatment plan
44. ~4 million datapoints molecularly-customized cancer treatment plan
45. ~4 million datapoints molecularly-customized cancer treatment plan Magic
46. Open
47. Open
48. Open
49. Source Open
50. Open
51. Data Open
52. Open
53. Access Open
54. Access
55. Access = Accessible?
56. OpenSource OpenData OpenAccess }Clear ?
61. Nature 483, 509 (29 March 2012) | doi:10.1038/483509a
62. Nature 483, 509 (29 March 2012) | doi:10.1038/483509a
63. the status quo tolerates poor communication of findings
64. the status quo tolerates poor communication of findings 6% 21% 8% 11% 54% cannot reproduce can reproduce in principle can reproduce w/ discrepancies can reproduce from processed data w/ discrepancies can reproduce partially
65. so what’s in principle reproducible, is not practically reproducible
66. 208,294,724 datapoints 124 pages supplemental material ? lines unobtainable source code ? version or architecture of R ? R packages key R package “ClaNC” no longer available 442 citations so what’s in principle reproducible, is not practically reproducible
67. define a question THE SCIENTIFIC PROCESS: it’s not one and done...
68. define a question gather information THE SCIENTIFIC PROCESS: it’s not one and done...
69. define a question gather information form a hypothesis THE SCIENTIFIC PROCESS: it’s not one and done...
70. define a question gather information form a hypothesis experimentally test your hypothesis THE SCIENTIFIC PROCESS: it’s not one and done...
71. define a question gather information form a hypothesis experimentally test your hypothesis analyze experimental data THE SCIENTIFIC PROCESS: it’s not one and done...
72. define a question gather information form a hypothesis experimentally test your hypothesis analyze experimental data draw conclusions THE SCIENTIFIC PROCESS: it’s not one and done...
73. define a question gather information form a hypothesis experimentally test your hypothesis analyze experimental data draw conclusions report, i.e. “publish” THE SCIENTIFIC PROCESS: it’s not one and done...
74. define a question gather information form a hypothesis experimentally test your hypothesis analyze experimental data draw conclusions report, i.e. “publish” validation by the scientific community validation by clinical trials THE SCIENTIFIC PROCESS: it’s not one and done...
76. generate data @ the bench or from a clinical cohort
77. generate data @ the bench or from a clinical cohort
78. generate data @ the bench or from a clinical cohort store on local server
79. generate data @ the bench or from a clinical cohort store on local server
80. analyze on local machine generate data @ the bench or from a clinical cohort store on local server
81. analyze on local machine generate data @ the bench or from a clinical cohort store on local server
82. analyze on local machine write a document generate data @ the bench or from a clinical cohort store on local server
83. submit to journal analyze on local machine write a document generate data @ the bench or from a clinical cohort store on local server
84. submit to journal analyze on local machine write a document sent to reviewers as pdf generate data @ the bench or from a clinical cohort store on local server
85. submit to journal analyze on local machine write a document sent to reviewers as pdf generate data @ the bench or from a clinical cohort store on local server
86. submit to journal analyze on local machine write a document sent to reviewers as pdf generate data @ the bench or from a clinical cohort accepted & digitally typeset store on local server
87. submit to journal analyze on local machine write a document sent to reviewers as pdf printed on paper generate data @ the bench or from a clinical cohort accepted & digitally typeset store on local server
88. submit to journal analyze on local machine write a document sent to reviewers as pdf printed on paper static html representation generate data @ the bench or from a clinical cohort accepted & digitally typeset store on local server
89. submit to journal analyze on local machine write a document sent to reviewers as pdf printed on paper static html representation generate data @ the bench or from a clinical cohort accepted & digitally typeset static pdf representation store on local server
90. submit to journal analyze on local machine write a document sent to reviewers as pdf printed on paper static html representation generate data @ the bench or from a clinical cohort accepted & digitally typeset static pdf representation store on local server
91. submit to journal analyze on local machine write a document sent to reviewers as pdf printed on paper static html representation generate data @ the bench or from a clinical cohort accepted & digitally typeset static pdf representation store on local server THE ACTUAL SCIENCE
92. submit to journal analyze on local machine write a document sent to reviewers as pdf printed on paper static html representation generate data @ the bench or from a clinical cohort accepted & digitally typeset static pdf representation store on local server THE ACTUAL SCIENCE SCIENTIFIC CLAIMS
93. analyze on local machine generate data @ the bench or from a clinical cohort store on local server THE ACTUAL SCIENCE
94. analyze on local machine generate data @ the bench or from a clinical cohort store on local server THE ACTUAL SCIENCE
96. clearScience
97. clearScience
98. clearScience
99. clearScience 1. re-imagine scientific communication
100. clearScience 1. re-imagine scientific communication 2. make every component of a data-intensive project (e.g. genomic analysis) web accessible
101. clearScience 1. re-imagine scientific communication 2. make every component of a data-intensive project (e.g. genomic analysis) web accessible 3. allow consumption of content at a variety of levels of complexity and abstraction leverage open, RESTful APIs 4.
102. leverage open, RESTful APIs
103. leverage open, RESTful APIs to let users reconstitute an entire analytical environment
104. leverage open, RESTful APIs hardware to let users reconstitute an entire analytical environment
105. leverage open, RESTful APIs hardware software to let users reconstitute an entire analytical environment
106. leverage open, RESTful APIs hardware software data to let users reconstitute an entire analytical environment
107. leverage open, RESTful APIs hardware software data code to let users reconstitute an entire analytical environment
108. hardware software data code
109. hardware software data code clearScience
110. hardware software data code clearScience a narrative that coherently interweaves scientific assertions with the executable data, code, software & hardware that support them
111. my computer’s the cloud, let me hand it to you” “
115. Betty Jones “I have a lump near my mastectomy scar”
117. josh greenberg brian bot stephen friend david burdick mike kellen myles axton editor nature genetics phil bourne founding editor-in-chief plos comp. biology ian mulvaney director of technology • mark patterson managing editor eLife sciences external partners kelly lamarco senior editor science trans. medicine eric schadt dir. multiscale mt. sinai school of med.
118. josh greenberg brian bot stephen friend david burdick mike kellen myles axton editor nature genetics phil bourne founding editor-in-chief plos comp. biology ian mulvaney director of technology • mark patterson managing editor eLife sciences external partners kelly lamarco senior editor science trans. medicine eric schadt dir. multiscale mt. sinai school of med. our patients