Data_description.txt

# Data Description

1. Processed data
There are two versions of preprocessed data generated.
Data files with the gene filtering process has suffix "_dat.RData".
Data files without the gene filtering process has suffix "_dat_raw.RData"
For datasets involved in Sim1 and Variance Analysis (GSE113590, GSE51984, GSE60424, GSE64655), we created "_dat_raw.RData", since all analysis is restricted to the same dataset.
For datasets involved in Sim2 and Sim3 (GSE51984, GSE60424, GSE64655, GSE115736, GSE118490), we created "_dat.RData", since the filtering process will be performed after data concatenation.
Thus, for datasets that involved in both groups (Sim1, Sim2 and Sim3), we created two versions of preprocessed data.
You can either process data from Raw quantification data or directly load processed data.
We recommend users to uncompress Processed data under the ./data folder to avoid path parameter reassignment.

2. Params Data
Params data contain basic settings of each simulation framework such as datasets, cellular component, factors, units and gene index (Details are in the Supplementary Table 3 of the publication)
3 sets of params are created for 3 simulation frameworks: Sim1, Sim2 and Sim3.
You can generate params data from "SimX"_params_set.R scripts.
Output corresponds to "SimX"_env.RData

3. Reference Data
Reference data contains reference used for deconvolution analysis.
Type of reference objects include: reference profile, averaged reference profile, signature gene sets, marker gene sets and reference sample annotation.
You can generate reference data from "SimX"_refGenerator.R scripts.
Output corresponds to "SimX"_ref.RData

4. Weight Data
Cellular component proportion data from 3 simulated frameworks.
sim1_W.txt, sim2_W_orthog.rds and sim2_W_real.rds

5. Mix Data
In silico mixture data of 3 simulated frameworks.
You can generate mixture data from "SimX"_mix.R scripts.
Output corresponds to "SimX"_mix.RData

6. Deconvolution Estimation Data
Estimation results of deconvolution analysis. 
You can generate deconvolution data from "SimX"_deconvolution_rPlatform.R and "SimX"_deconvolution_crossPlatform.R scripts.
Output corresponds to "SimX"_estW.RData 

7. Other Data
sim1_diagnostics_simModel.RData are variance analysis for Sim1_simModel.
You can generate diagnostic data from "sim1_diagnostics_simModel.R"