This package is in no way officially related to or endorsed by the PPMI.
In the field of Parkinson’s disease (PD) therapeutics, the ultimate goal is to develop disease-modifying treatments that slow, prevent or reverse the underlying disease process. Validated biomarkers of disease progression would dramatically accelerate PD therapeutics research. Current progression biomarkers, however, are not optimal and are not fully validated. (source: PPMI website)
PPMI (Parkinson's Progression Markers Initiative) is an observational clinical study to verify progression markers in Parkinson’s disease. PPMI has emerged as a model for following multiple cohorts of significant interest and is being conducted at a network of clinical sites around the world. The study is designed to establish a comprehensive set of clinical, imaging and biosample data that will be used to define biomarkers of PD progression. Once these biomarkers are defined, they can be used in therapeutic studies, which is the ultimate goal. (source: PPMI website)
PPMI will follow standardized data acquisition protocols to ensure that tests and assessments conducted at multiple sites and across multiple cohorts can be pooled in centralized databases and repositories. The clinical, imaging and biologic data will be easily accessible to researchers in real time through the PPMI website. The biological samples collected throughout the course of PPMI will be stored in a central repository that will be accessible to any scientist with promising biomarker leads for the purposes of verifying initial results and assessing correlations to clinical outcomes and other biomarkers. (source: PPMI website)
Data are dumped from the PPMI repository. They are then preprocessed, and derived variables are computed as recommended by the PPMI.
The following files are parsed using the D01-PPMI-Parsing.Rmd script.
- Benton_Judgment_of_Line_Orientation.csv
- Blood_Chemistry___Hematology.csv
- Center-Subject_List.csv
- Clinical_Diagnosis_and_Management.csv
- Code_List.csv
- Cognitive_Categorization.csv
- Concomitant_Medications.csv
- Current_Biospecimen_Analysis_Results.csv
- Data_Dictionary.csv
- DATScan_Analysis.csv
- DaTscan_Imaging.csv
- Diagnostic_Features.csv
- Epworth_Sleepiness_Scale.csv
- Family_History__PD_.csv
- General_Neurological_Exam.csv
- General_Physical_Exam.csv
- Genetic_Testing_Results.csv
- Geriatric_Depression_Scale__Short_.csv
- Hopkins_Verbal_Learning_Test.csv
- Letter_-_Number_Sequencing__PD_.csv
- Magnetic_Resonance_Imaging.csv
- MDS_UPDRS_Part_I__Patient_Questionnaire.csv
- MDS_UPDRS_Part_I.csv
- MDS_UPDRS_Part_II__Patient_Questionnaire.csv
- MDS_UPDRS_Part_III.csv
- MDS_UPDRS_Part_IV.csv
- Modified_Schwab_+_England_ADL.csv
- Montreal_Cognitive_Assessment__MoCA_.csv
- Neurological_Exam_-_Cranial_Nerves.csv
- Patient_Status.csv
- PD_Features.csv
- PPMI_PD_Variants_Genetic_Status_WGS_20180921.csv
- Primary_Diagnosis.csv
- QUIP_Current_Short.csv
- REM_Sleep_Disorder_Questionnaire.csv
- SCOPA-AUT.csv
- Screening___Demographics.csv
- Semantic_Fluency.csv
- Socio-Economics.csv
- State-Trait_Anxiety_Inventory.csv
- Symbol_Digit_Modalities.csv
- TAP-PD_Kinetics_Device_Testing.csv
- University_of_Pennsylvania_Smell_ID_Test.csv
- Use_of_PD_Medication.csv
- Vital_Signs.csv
The following objects are created and saved in the PPMI-Parsed.rda file.
- `dumpDate`
- `visitDoc`
- `visitInfo`
- `scheduledVisits`
- `varDoc`
- `patientData`
- `visitData`
This document describes how PPMI variables were derived from the original parsed data.
Derived variables are computed using the D02-PPMI-Variables.Rmd script as described in the Derived_Variable_Definitions_and_Score_Calculations.csv
file.
The former objects are updated and saved in the PPMI-Derived.rda file.
Genotyping data were downloaded and the IMMUNO SNP coordinates were converted into GRCh37 before merging with the NEUROX SNPs using the plink software:
plink --bfile IMMUNO-GRCh37 --bmerge NEUROX.bed NEUROX.bim NEUROX.fam --make-bed --out ImmunoNeurox-GRCh37
## - mind:0.6 because the 2 arrays were performed on different individuals:
## 0.6 corresponds approximatevely to the proportion of SNPs on one array
## compared to the sum of the 2 arrays
## geno:0.2 because the 2 arrays were performed on different individuals:
## 0.2 corresponds approximatevely to the proportion of individuals
## on one array compared to the sum of the 2 arrays
plink --bfile ImmunoNeurox-GRCh37 --mind 0.6 --maf 0.05 --geno 0.2 --hwe 0.001 --make-bed --out ImmunoNeurox-GRCh37-Filtered
The GRCh37 coordinates for the IMMUNO SNPs are available in the Genotyping/IMMUNO-GRCh37.bim file. After download, the files IMMUNO.bed and IMMUNO.fam need to be renamed IMMUNO-GRCh37.bed and IMMUNO-GRCh37.fam respectively before merging.
The PPMI R package is generated using the S01-PackageGenerator.R script. This script also builds the "PPMI-Data-Usage" vignette using updated data.
After building and installing the PPMI R package, data can be exported in an Excel file using the Data-Summary-Export.Rmd script.
After building and installing the PPMI R package, data can be exported in
an SAS XPORT file using the Export-SAS.R script. However the function
from the SASxport package (write.xport()) used to build this file raises
several messages about truncated long names. Also in the PPMI R package,
row names are used to make cross-references between data frames. I don't
know has row names are exported by SASxport and how they are handled by
SAS itself. CSV files are also produced by this script.