acrl trust in science talk

44
Is there a role for trust in science? March 31, 2011 Association of College and Research Libraries Jean-Claude Bradley Department of Chemistry Drexel University

Upload: jean-claude-bradley

Post on 11-May-2015

1.478 views

Category:

Education


0 download

DESCRIPTION

Jean-Claude Bradley presents at the Association of College and Research Libraries on March 31, 2011 on "Is there a role for Trust in Science".

TRANSCRIPT

Page 1: ACRL Trust in Science Talk

Is there a role for trust in science?

March 31, 2011

Association of College and Research Libraries

Jean-Claude Bradley

Department of ChemistryDrexel University

Page 2: ACRL Trust in Science Talk

The Chemical Information Validation Sheet

567 curated and referenced measurements from Fall 2010 Chemical Information Retrieval course

Page 3: ACRL Trust in Science Talk

The Chemical Information Validation Explorer

(Andrew Lang)

Page 4: ACRL Trust in Science Talk

Discovering outliers for melting points (stdev/average)

Page 5: ACRL Trust in Science Talk

Investigating the m.p. inconsistencies of EGCG

Page 6: ACRL Trust in Science Talk

Investigating the m.p. inconsistencies of cyclohexanone

Page 7: ACRL Trust in Science Talk

Sigma-Aldrich, Acros and Wolfram Alpha apparently use the same sources for melting

points

Page 8: ACRL Trust in Science Talk

Sigma-Aldrich, Acros and Wolfram Alpha apparently use the same sources for boiling

points

Page 9: ACRL Trust in Science Talk

Sigma-Aldrich, Acros and Wolfram Alpha apparently

DO NOT use the same sources for flash points

Page 10: ACRL Trust in Science Talk

Most popular data sources

Page 11: ACRL Trust in Science Talk

Alfa Aesar donates melting points to the public

Page 12: ACRL Trust in Science Talk

Open Melting Point Explorer

Page 13: ACRL Trust in Science Talk

Outliers

MDPI dataset

EPI (via ChemSpider)

Page 14: ACRL Trust in Science Talk

Outliers

Alfa Aesar

Page 15: ACRL Trust in Science Talk

Inconsistencies and SMILES problems within MDPI dataset

Page 16: ACRL Trust in Science Talk

MDPI Dataset labeled with High Trust Level

Page 17: ACRL Trust in Science Talk

Open Melting Point Datasets

Page 18: ACRL Trust in Science Talk

Open Random Forest modeling of Open Melting Point data using CDK descriptors

(Andrew Lang)

R2 = 0.78, TPSA and nHdon most important

Page 19: ACRL Trust in Science Talk

Melting point prediction service

Page 20: ACRL Trust in Science Talk

Using melting point for temperature dependent solubility prediction

Page 21: ACRL Trust in Science Talk

Motivation: Faster Science, Better Science

Page 22: ACRL Trust in Science Talk

There are NO FACTS, only measurements embedded

within assumptions

Open Notebook Science maintains the integrity of data

provenance by making assumptions explicit

Page 23: ACRL Trust in Science Talk

TRUST

PROOF

Page 24: ACRL Trust in Science Talk

First record then abstract structure

In order to be discoverable use Google friendly formats (simple HTML, no login)

In order to be replicable use free hosted tools (Wikispaces, Google Spreadsheets)

Strategy for an Open Notebook:

Page 25: ACRL Trust in Science Talk

Crowdsourcing Solubility Data

Page 26: ACRL Trust in Science Talk

Data provenance: From Wikipedia to…

Page 27: ACRL Trust in Science Talk

…the lab notebook and raw data

Page 28: ACRL Trust in Science Talk

The importance of raw data availability

Missed in a prior publication on solubility

for this compound

Page 29: ACRL Trust in Science Talk

Solubilities collected in a Google Spreadsheet

Page 30: ACRL Trust in Science Talk

Rajarshi Guha’s Live Web Query using Google Viz API

Page 31: ACRL Trust in Science Talk

Web services for summary data

(Andrew Lang)

Page 32: ACRL Trust in Science Talk

Web service calls from within a Google Spreadsheet for solubility measurement and

prediction

(Andrew Lang)

Page 33: ACRL Trust in Science Talk

Integration of Multiple Web Services to Recommend Solvents for Reactions

(Andrew Lang)

Page 34: ACRL Trust in Science Talk
Page 35: ACRL Trust in Science Talk
Page 36: ACRL Trust in Science Talk
Page 37: ACRL Trust in Science Talk

Reaction Attempts Book

Page 38: ACRL Trust in Science Talk

Reaction Attempts Book: Reactants listed Alphabetically

Page 39: ACRL Trust in Science Talk

ONS Challenge Solubility Book cited for nanotechnology application

Page 40: ACRL Trust in Science Talk

Lulu.com Data Disks

Page 41: ACRL Trust in Science Talk

All ONS web services

Page 42: ACRL Trust in Science Talk

Dynamic links to private tagged Mendeley collections

(Andrew Lang)

Page 43: ACRL Trust in Science Talk

For all Formats of ONS Projects

Page 44: ACRL Trust in Science Talk

Conclusions

•Trust has no role in science and is unnecessary if sufficient proof is provided

•The peer-reviewed system is critical but insufficient to communicate all available actionable scientific information