Simpati: Repository of Supplementary data and methods

CONTENT

• Data and methods to replicate the results obtained by Simpati, netDx and PASNet

REQUIREMENTS

1. Install Docker for your operating system following the link
2. Pull two containers (one containing Simpati enviroment, one netDx and one PASNet): lgiudice/simpati:latest and lgiudice/comp1:latest and lgiudice/comp2:latest
   • With bash:

   docker pull lgiudice/simpati:latest
   docker pull lgiudice/comp1:latest
   docker pull lgiudice/comp2:latest
   

PIPELINE TO GET Simpati RESULTS AND PERFORMANCES ON RNAseq AND SOMATIC MUTATION DATA

1. Download and unzip the Simpati data and scripts from the link
   • With bash: wget https://univr-my.sharepoint.com/:u:/g/personal/luca_giudice_univr_it/EeQJQb335QlGtTdMd3ni2mIB79wUeVy151SpVG8ULWqi2w?download=1 --no-check-certificate -O Simpati_v31_stable.7z
2. Simpati_v31_stable/output/op1 directory contains the results used in the publication, if you want to get fresh new output data you can remove it
3. Run the simpati container building the unzipped Simpati directory to the following path: /home/lgiudice/containers/Simpati_v31_stable
   • With bash: docker run --rm -it -v <full_user_path>/Simpati_v31_stable:/home/lgiudice/containers/Simpati_v31_stable lgiudice/simpati:latest /bin/bash
4. Modify the script op1_Classification_v9_Mut_par.R at the line 91 and 96 based on your computational resources, setting the RAM memory and the number of cores
5. Modify the script op1_Classification_v9_RNAseq_par.R at the line 91 and 96 based on your computational resources, setting the RAM memory and the number of cores
6. Now you are ready to run Simpati:

   Rscript op1_Classification_v9_Mut_par.R
   Rscript op1_Classification_v9_RNAseq_par.R
   

PIPELINE TO GET netDx RESULTS AND PERFORMANCES ON SOMATIC MUTATION DATA

1. Download and unzip the netDx data and scripts from the link
   • With bash: wget https://univr-my.sharepoint.com/:u:/g/personal/luca_giudice_univr_it/EdeJn9yyrfFNv0rAikZF1gYBdoLzM_9HGWU8Pyw7wVJNhQ?download=1 --no-check-certificate -O netDx_Mut.7z
2. netDx/output directory contains the results used in the publication, if you want to get fresh new output data you can remove everything except: op2_TCGA4netDx_mut
3. Run the comp1 container building the unzipped netDx directory to the following path: /home/lgiudice/containers/netDx_Mut
   • With bash: docker run --rm -it -v <full_user_path>/netDx_Mut:/home/lgiudice/containers/netDx_Mut lgiudice/comp1:latest /bin/bash
4. Modify the script op3_multi_omics_TCGA.R at the line 101 based on your computational resources, setting the RAM memory and the number of cores
5. Modify the script op3_multi_omics_TCGA_large.R at the line 101 based on your computational resources, setting the RAM memory and the number of cores
6. Now you are ready to run netDx:

   Rscript op3_multi_omics_TCGA.R
   Rscript op3_multi_omics_TCGA_large.R
   

PIPELINE TO GET netDx RESULTS AND PERFORMANCES ON RNAseq DATA

1. Download and unzip the netDx data and scripts from the following link
   • With bash: wget https://univr-my.sharepoint.com/:u:/g/personal/luca_giudice_univr_it/EXaGdLX9_e5Huj_KjoOfF4gB_HcOou9ghoTQ1zF0ZI77zw?download=1 --no-check-certificate -O netDx_RNAseq.7z
2. netDx/output directory contains the results used in the publication, if you want to get fresh new output data you can remove everything except: op2_TCGA4netDx_RNAseq
3. Run the comp1 container building the unzipped netDx directory to the following path: /home/lgiudice/containers/netDx_RNAseq
   • With bash: docker run --rm -it -v <full_user_path>/netDx_RNAseq:/home/lgiudice/containers/netDx_RNAseq lgiudice/comp1:latest /bin/bash
4. Modify the script op3_multi_omics_TCGA.R at the line 68 based on your computational resources, setting the RAM memory and the number of cores
5. Modify the script op3_multi_omics_TCGA_large.R at the line 68 based on your computational resources, setting the RAM memory and the number of cores
6. Now you are ready to run netDx

   Rscript op3_multi_omics_TCGA.R
   Rscript op3_multi_omics_TCGA_large.R
   

PIPELINE TO GET PASNet RESULTS AND PERFORMANCES ON RNAseq DATA

1. Download and unzip the PASNet data and scripts from the following link
2. PASNet/output//output_py directory contains the results used in the publication for a specific dataset, if you want to get fresh new output data you can delete that directory
3. Run the comp2 container building the unzipped PASNet directory to the following path: /home/PASNet
   • With bash: docker run --rm -it -v <full_user_path>/PASNet:/home/PASNet lgiudice/comp2:latest /bin/bash
4. Now you are ready to run PASNet workflow: first you will tune the model and then run the model with the tuned parameters

   python op1_Run_EmpiricalSearch
   python op2_Run
   

Links to the original softwares

• netDx
• PASNet