jasenzhang1/cse284_group_16

CSE 284 Group 16 Project

CSE284_Group_16

TCGA Website Link: https://xenabrowser.net/datapages/?cohort=GDC%20TCGA%20Lung%20Squamous%20Cell%20Carcinoma%20(LUSC)&removeHub=https%3A%2F%2Fxena.treehouse.gi.ucsc.edu%3A443

Gene data: “HTSeq - Counts”

https://xenabrowser.net/datapages/?dataset=TCGA-LUSC.htseq_counts.tsv&host=https%3A%2F%2Fgdc.xenahubs.net&removeHub=https%3A%2F%2Fxena.treehouse.gi.ucsc.edu%3A443

Phenotype data: “phenotype”

https://xenabrowser.net/datapages/?dataset=TCGA-LUSC.GDC_phenotype.tsv&host=https%3A%2F%2Fgdc.xenahubs.net&removeHub=https%3A%2F%2Fxena.treehouse.gi.ucsc.edu%3A443

Survival data: “survival”

https://xenabrowser.net/datapages/?dataset=TCGA-LUSC.survival.tsv&host=https%3A%2F%2Fgdc.xenahubs.net&removeHub=https%3A%2F%2Fxena.treehouse.gi.ucsc.edu%3A443

Data dictionary: https://docs.gdc.cancer.gov/Data_Dictionary/viewer

From the TCGA websites, download:

• TCGA-LUSC.GDC_phenotype.tsv
• TCGA-LUSC.htseq_counts.tsv
• TCGA-LUSC.survival.tsv and unzip the first two gz files.

Code documentation:

• workspace Reads in the raw downloaded files and prepares df_train and df_test RData files as well as the annotations Excel file.

• workspace2 Reads in the df_test and df_train RData files and generates uninformative, OLS, and LMM models and figures.

• ryo_analysis Explores data and methods of specifying random effects in LMMs. Compares LMMs with differing covariates and error terms. Generates Manhattan plots for protein-coding genes.

• formulas Formats Null, OLS, and LMM formulas for slides and report.