Improve Research Reproducibility A Bio-protocol resource
bpdetail_icon_lock
Preprint

Statistical analysis

DC David Chen
JW Josephine Wu
GV Gordana Vunjak-Novakovic

Last updated date: May 9, 2021 Views: 873 Forks: 0

original An abbreviated version of this protocol was published in Science Translational Medicine in Oct, 2020
Tissue engineered autologous cartilage-bone grafts for temporomandibular joint regeneration
download pdf Download PDF
ask Ask a question
cite How to cite
nfavorite Favorite
  1. Open GraphPad Prism software. This protocol was written based on GraphPad Prism 8.
  2. Create a data table.
    NEW TABLE & GRAPH > Grouped > Data table: Enter or import data into a new table > Options: Enter and plot a single Y value for each point > Create
  3. Enter group names (e.g. Native, Cartilage-bone, Bone-only, Acellular) and data for a single assay. This data table will be referred to as Data X in the subsequent steps. For each assay, create a separate data table.
    Data Tables: + New Data Table…
  4. Native data, which were expected to be normally distributed, were checked for outliers using the robust regression and outlier removal method and a Q coefficient of 1% to limit the false discovery rate.
    Data Tables: Data X > Analysis: Analyze > Column analyses: Identify outliers > Analyze which data sets? Native > OK > Method: ROUT (recommended; can find any number of outliers) > How aggressive? Q = 1% > OK
  5. View outliers within the native data.
    Results: Identify outliers of Data X > Outliers
  6. Remove outliers within the native data.
    Data Tables: Data X > remove values identified in previous step.
  7. If n>4 in each group, check for whether they fit Gaussian distributions using the Shapiro-Wilk normality test.
    Data Tables: Data X > Analysis: Analyze > Column analyses: Normality and Lognormality Tests > Analyze which data sets? Select all > OK > Which distribution(s?) to test? Normal (Gaussian distribution) > Methods to test distributions(s): Shapiro-Wilk normality test > Calculations: Significance level (alpha): 0.05 > OK
  8. View results of normality test.
    Results: Normality and Lognormality Tests of Data X > Passed normality test (alpha=0.05)?
  9. If n≥4 and normality test was passed, carry out statistical analysis by one-way analysis of variance (ANOVA) test and post hoc Tukey-Kramer’s comparison test to compare means between groups.
    Data Tables: Data X > Analysis: Analyze > Column analyses: One-way ANOVA (and nonparametric or mixed) > Analyze which data sets? Select all > OK > Experimental design: No matching or pairing > Assume Gaussian distribution? Yes. Use ANOVA. > Assume equal SDs? Yes. Use ordinary ANOVA test. > Multiple Comparisons > Followup tests: Compare the mean of each column with the mean of every other column. > Options > Multiple comparisons test: Correct for multiple comparisons using statistical hypothesis testing. Recommended. > Test: Tukey (recommended). > Multiple comparisons options: Report multiplicity adjusted P value for each comparison. Family-wise alpha threshold and confidence level: 0.05 (95% confidence interval) > OK
  10. If n<4 or normality test was not passed, carry out statistical analysis by nonparametric Kruskal-Wallis test and post hoc Dunn’s multiple comparison test to compare the means between groups.
    Data Tables: Data X > Analysis: Analyze > Column analyses: One-way ANOVA (and nonparametric or mixed) > Analyze which data sets? Select all > OK > Experimental design: No matching or pairing > Assume Gaussian distribution? No. Use Nonparametric test. > Multiple Comparisons > Followup tests: Compare the mean rank of each column with the mean rank of every other column. > Options > Multiple comparisons test: Correct for multiple comparisons using statistical hypothesis testing. Recommended. Test: Dunn’s. > Multiple comparisons options: Report multiplicity adjusted P value for each comparison. Family-wise alpha threshold and confidence level: 0.05 (95% confidence interval) > OK
  11. View results of ANOVA or Kruskal-Wallis tests.
    Results > Ordinary one-way ANOVA of Data X or Kruskal-Wallis test of Data X > Multiple comparisons > Summary
Copyright: Content may be subjected to copyright.
How to cite：
Readers should cite both the Bio-protocol preprint and the original research article where this protocol was used:
  1. Chen, D, Wu, J and Vunjak-Novakovic, G(2021). Statistical analysis. Bio-protocol Preprint. bio-protocol.org/prep1068.
  2. Chen, D., Wu, J. Y., Kennedy, K. M., Yeager, K., Bernhard, J. C., Ng, J. J., Zimmerman, B. K., Robinson, S., Durney, K. M., Shaeffer, C., Vila, O. F., Takawira, C., Gimble, J. M., Guo, X. E., Ateshian, G. A., Lopez, M. J., Eisig, S. B. and Vunjak-Novakovic, G.(2020). Tissue engineered autologous cartilage-bone grafts for temporomandibular joint regeneration . Science Translational Medicine 12(565). DOI: 10.1126/scitranslmed.abb6683

Do you have any questions about this protocol?

Post your question to gather feedback from the community. We will also invite the authors of this article to respond.

post Post a Question
0 Q&A
This protocol preprint was submitted via the "Request a Protocol" track.