Experimental design and data analysis

From OpenWetWare

(Difference between revisions)
Jump to: navigation, search
(2. Planning your project)
Current revision (05:03, 14 April 2014) (view source)
(4. Psychological pitfalls)
 
(12 intermediate revisions not shown.)
Line 8: Line 8:
== 1. Selecting a project ==
== 1. Selecting a project ==
 +
* [http://www.ncbi.nlm.nih.gov/pubmed/19782018 How to choose a good scientific problem.] - 2009 review by Alon U in Mol. Cell
== 2. Planning your project ==
== 2. Planning your project ==
-
* [http://www.ingeniumweb.com/blog/post/top-five-best-free-online-project-management-application-services/1027/ 5 free project planning software suites] - software review by Ingenium Llc., a web design company, from 2013
+
====Background check, methods research====
 +
* [http://www.ncbi.nlm.nih.gov/pubmed/ PubMed] - your classical literature database; "weak in some areas of chemistry, physics, plant science, maths & stats" ~ j
 +
* [http://scholar.google.de/ Google Scholar] - Google's scientific material database; "pro: citation count, PDFs, con: order of articles not disclosed and older articles often on top" ~ j
 +
 
 +
* [[Image:Padlock-closed-S.png]] [http://www.currentprotocols.com/WileyCDA/ Current Protocols] - life science protocols, only paid access
 +
* [[Image:Padlock-closed-S.png]] [http://cshprotocols.cshlp.org/ Cold Spring Harbor Protocols] - protocols for subscribers, also publisher of the widespread Molecular Cloning book series
 +
* [[Image:Padlock-closed-S.png]] [http://www.jove.com/ JoVE] - Journal of visualized experiments, some teaser videos available; see also OWW article on [[JoVE]]
 +
 
 +
====Discussion forums====
 +
* [http://www.protocol-online.org/ Protocol Online] and its subsection [http://www.protocol-online.org/forums/index.php BioForum]
 +
* [http://molecularbiology.forums.biotechniques.com/ BioTechniques] - techniques, tips, and questions concerning molecular and cell biology
 +
* [http://seqanswers.com/ SEQanswers] - deep sequencing discussion forum
 +
 
 +
====Other====
 +
* [http://www.ingeniumweb.com/blog/post/top-five-best-free-online-project-management-application-services/1027/ 5 free project planning software suites] - 2013 software review by Ingenium Llc., a web design company
== 3. Data analysis ==
== 3. Data analysis ==
* [http://www.methodenberatung.uzh.ch/datenanalyse.html overview diagram to decide which statistical test to use] - in German by the University of Zürich
* [http://www.methodenberatung.uzh.ch/datenanalyse.html overview diagram to decide which statistical test to use] - in German by the University of Zürich
 +
* [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1924630/ A clinician-educator's roadmap to choosing and interpreting statistical tests] - 2006 review
* [http://udel.edu/~mcdonald/statintro.html Handbook of biological statistics] - online textbook by John McDonald at U Delaware
* [http://udel.edu/~mcdonald/statintro.html Handbook of biological statistics] - online textbook by John McDonald at U Delaware
 +
* [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2550545/ Absence of evidence is not evidence of absence] - 1995 article by Altman and Bland, part of an [[BMJ Statistics Notes series|excellent series on statistics]]
 +
* [http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2064100/ Error bars in experimental biology] - 2007 article by Cummings and colleagues
* [http://rfd.uoregon.edu/files/rfd/StatisticalResources/outl.txt Dealing with outliers] - very detailed essay on the topic
* [http://rfd.uoregon.edu/files/rfd/StatisticalResources/outl.txt Dealing with outliers] - very detailed essay on the topic
-
 
+
====Software====
 +
* [[Guide to statistics software]]
* [http://www.youtube.com/watch?v=iRrcUzHF-rE video tutorial: non-parametric rank sum test with Excel] - see how a basic rank sum test is done using Excel
* [http://www.youtube.com/watch?v=iRrcUzHF-rE video tutorial: non-parametric rank sum test with Excel] - see how a basic rank sum test is done using Excel
* [https://www.gnu.org/software/octave/ free Matlab alternative Octave]
* [https://www.gnu.org/software/octave/ free Matlab alternative Octave]
== 4. Psychological pitfalls ==
== 4. Psychological pitfalls ==
 +
* [http://www.ted.com/talks/angela_lee_duckworth_the_key_to_success_grit Angela L Duckworth: Grit] - 2013 TED talk
 +
* [http://www.ted.com/talks/derek_sivers_keep_your_goals_to_yourself Derek Silvers: Keep your goals to yourself] - 2010 TED talk
 +
* Dan Ariely on procrastination and deadlines in [http://danariely.com/?s=Magic+of+Procrastination students] and [http://on.wsj.com/1epdijX professors]
 +
 +
 +
* [http://www.newyorker.com/reporting/2010/12/13/101213fa_fact_lehrer The truth wears off] - 2010 magazine article by Lehrer on the often declining effect of phenomena over time; see also the [https://en.wikipedia.org/wiki/Decline_effect Wikipedia]
 +
* [http://www.ncbi.nlm.nih.gov/pubmed/18180760 How effective are second-generation antipsychotic drugs?] - 2009 meta analysis of schizophrenia drugs shows a declining effect with time
 +
* Share your negative results at [http://bit.ly/1epf58H PLoS ONE] and the [http://www.jnrbm.com/ Journal of Negative Results in Biomedicine]
 +
 +
 +
* [http://www.ncbi.nlm.nih.gov/pubmed/12153921 Association between competing interests and authors' conclusions] - Kjaergard, Als-Nielsen 2002
 +
: "Authors' conclusions.. significantly favoured experimental interventions if financial competing interests were declared." from the conclusion
 +
* [http://link.springer.com/article/10.1007/BF02602327 Source of funding and outcome of clinical trials] - Davidson 1986;
 +
:"90% of privately supported but only 60% of generally supported clinical studies reach the conclusion that the new therapy is better." ~ j
-
== Organizer & host institution ==
+
== See also ==
 +
* OpenWetWare [[Courses]]
 +
* [https://ovidius.uni-tuebingen.de/ilias3/goto.php?target=crs_765899&client_id=pr02 2014 ILIAS course page] (requires login)
 +
* [https://ovidius.uni-tuebingen.de/ilias3/goto.php?target=crs_567697&client_id=pr02 2013 ILIAS course page] (requires login)

Current revision

This page lists resources discussed in the Design and Analysis seminar and includes links to relevant further reading. Please feel free to add your own suggestions and comments to the sections. The course is run approximately yearly and takes places in the Institute of Biochemistry of the University of Tübingen. See the Institute's course page for dates and contact information.

Aim of the course

Experimental design and data analysis is a new graduate seminar piloted in 2013 to address the questions of how to plan an experiment and how to best analyze the resulting data. We look at how to do a proper background check, where to find the best protocols, how to formulate a useful hypothesis, methods to keep experiments on schedule, tools of data analysis, and finally we will talk about some psychological pitfalls frequently seen in the interpretation of results.

1. Selecting a project

2. Planning your project

Background check, methods research

  • PubMed - your classical literature database; "weak in some areas of chemistry, physics, plant science, maths & stats" ~ j
  • Google Scholar - Google's scientific material database; "pro: citation count, PDFs, con: order of articles not disclosed and older articles often on top" ~ j
  • Image:Padlock-closed-S.png Current Protocols - life science protocols, only paid access
  • Image:Padlock-closed-S.png Cold Spring Harbor Protocols - protocols for subscribers, also publisher of the widespread Molecular Cloning book series
  • Image:Padlock-closed-S.png JoVE - Journal of visualized experiments, some teaser videos available; see also OWW article on JoVE

Discussion forums

Other

3. Data analysis

Software

4. Psychological pitfalls



"Authors' conclusions.. significantly favoured experimental interventions if financial competing interests were declared." from the conclusion
"90% of privately supported but only 60% of generally supported clinical studies reach the conclusion that the new therapy is better." ~ j

See also

Personal tools