The rodent object-in-context task: a systematic review and meta-analysis of important variables

Systematic Review and meta-analysis as described in:

• Milou S.C. Sep, Marijn Vellinga, R. Angela Sarabdjitsingh, Marian Joëls. (2021). Measuring context-dependent memory in rodents: a systematic review and meta-analysis of important variables in the object-in-context task. bioRxiv 2021.03.12.435070; doi: https://doi.org/10.1101/2021.03.12.435070 [preprint]
• Milou S.C. Sep, Marijn Vellinga, R. Angela Sarabdjitsingh, Marian Joëls. (in press). The rodent object-in-context task: a systematic review and meta-analysis of important variables. PLOS ONE.

Index

_ README.md: an overview of the project
|___ data: data files used in the project
|___ processed_data: intermediate files from the analysis
|___ results: results of the analyses (data, tables, figures)
|___ R: contains all R-code in the project

1. Flowchart

• script: flowchart.R
• input (script contains code that retrieves data from OSF):
  • search step:
    • data/hits.search.thesis.MV.txt
    • data/hits.new.search.meta.oic.v25.5.20.txt
  • screening step:
    • data/Screening S1 thesis search PMIDs.csv
    • data/Screening S2 new.in.new.search.PMIDs.csv
  • included data:
    • data/280121_Data_Extraction_RoB.xlsx
• actions:
  • count numbers screening & inclusions for flow chart

2. Preprocessing

• script: prepare_data.Rmd (for Systematic review & meta-analysis)
• input: data/280121_Data_Extraction_RoB.xlsx (script contains code that retrieves data from OSF)
• actions:
  • pre-processing data (and save cleaned data for Systematic review table)
  • missing values.
    1. Variables with more than 1/3 missing are excluded from Random forest-based meta-analysis
    2. Other variables: Missing values are replaced by median value (for numeric) of most prevalent category (for factors)
  • create sum scores: Arousal.Prior, Context.Difference.Score, Arousal.Task.Habituation, Arousal.Total
• output:
  • processed_data/SR_data.RDS (for Systematic review table)
  • processed_data/cleaned_data.RDS (for meta-analysis)

3. Systematic review Table

• script: systematic_review_table.Rmd
• action: create systematic review table
• output: results/Overview_SR.docx

4. Visualize Study Quality and Risk of Bias

• script: QA_RoB_plots.Rmd
• action: create waffel plot with SYRCLE’s risk of bias assessment per study (PMID)
• output: results/QA_ROB.tiff

5. Random-effects meta-analysis

• script: random_effects_meta_analysis.Rmd
• input:
  • processed_data/cleaned_data.RDS
  • data/280121_Data_Extraction_RoB.xlsx
• actions:
  • random-effects meta-analysis
  • calculate required sample size for future studies
  • robustness of effects measures
  • sensitivity analyses
• output:
  • processed_data/data_with_effect_size.RDS
  • results/forest_year.tiff
  • results/funnel.colours.tiff
  • results/study.quality.jpeg

6. Random forest-based meta-analysis

• script: MetaForest.Rmd
• Input: processed_data/data_with_effect_size.RDS
• Actions:
  • tune & run random forest-based meta-analysis (MetaForest)
  • identify important moderators based on variable importance in RF
• Output:
  • processed data:
    • processed_data/datForest_for_WS_Plot.RDS
    • processed_data/fitted.MetaForest.RDS
    • processed_data/important_variables.RDS
  • results:
    • results/metaforest_convergencePlot.jpeg
    • results/metaforest_varImportance.jpeg
    • results/important_variables_metaforest.csv

7. MetaForest follow-up: Partial dependence and Weighted scatter plots

• script: MetaForest_PD_WS_plots.Rmd
• input:
  • processed_data/fitted.MetaForest.RDS
  • processed_data/datForest_for_WS_Plot.RDS
  • processed_data/important_variables.RDS
• actions:
  • create partial dependance (PD) and weighted scatter plots to follow-up most important variables MetaForest
• output:
  • results/metaforest_adapted_PD_plots.jpeg
  • results/metaforest_adapted_WS plots.jpeg