analysis_plan.Rmd

---
title: "QGT-Columbia-analysis-plan"
author: "Hae Kyung Im"
date: "2020-06-03"
output: workflowr::wflow_html
editor_options:
  chunk_output_type: console
---

# Preliminary information
Data and copies of repositories can be downloaded from [Box here](https://uchicago.box.com/s/zhapf2zfxcpj7thvq4sjnqale3emleum)

The latest version of the analysis plan is on [github here](https://github.com/hakyimlab/QGT-Columbia-HKI/blob/master/analysis/analysis_plan.Rmd)

# Transcriptome-wide association methods

```{r testing}

print(getwd())

```

## predict expression 

```{bash predict genetic component of expression,eval=FALSE}

export DATA=$PRE/s-predixcan/data
export MODEL=$PRE/models
export METAXCAN=$PRE/MetaXcan-master/software
export OUTPUT=$PRE/results

printf "Predict expression\n\n"
python3 $METAXCAN/Predict.py \
--model_db_path $DATA/models/gtex_v8_en/en_Whole_Blood.db \
--vcf_genotypes $DATA/1000G_hg38/ALL.chr22.shapeit2_integrated_snvindels_v2a_27022019.GRCh38.phased.vcf.gz \
--vcf_mode genotyped \
--variant_mapping $DATA/gtex_v8_eur_filtered_maf0.01_monoallelic_variants.txt.gz id rsid \
--on_the_fly_mapping METADATA "chr{}_{}_{}_{}_b38" \
--prediction_output $RESULTS/vcf_1000G_hg38_en/Whole_Blood__predict.txt \
--prediction_summary_output $RESULTS/vcf_1000G_hg38_en/Whole_Blood__summary.txt \
--verbosity 9 \
--throw

printf "association\n\n"
python3 $METAXCAN/PrediXcanAssociation.py \
--expression_file $RESULTS/vcf_1000G_hg38_en/Whole_Blood__predict.txt \
--input_phenos_file $DATA/1000G_hg38/random_pheno_1000G_hg38.txt \
--input_phenos_column pheno \
--output $RESULTS/vcf_1000G_hg38_en/Whole_Blood__association.txt \
--verbosity 9 \
--throw

```

# Summary PrediXcan

## download harmonized and imputed GWAS result for coronary artery disease

```{r}



```


```{bash, eval=TRUE}

export PRE=/Users/haekyungim/Box/LargeFiles/QGT-Columbia-HKI
export DATA=$PRE/s-predixcan/data
export MODEL=$PRE/models
export METAXCAN=$PRE/MetaXcan-master/software
export OUTPUT=$PRE/results

echo $PRE
echo $DATA
echo $MODEL
echo $METAXCAN
echo $OUTPUT

```

## run s-predixcan

```{bash, eval=FALSE}

export PRE=/Users/haekyungim/Box/LargeFiles/QGT-Columbia-HKI
export DATA=$PRE/s-predixcan/data
export MODEL=$PRE/models
export METAXCAN=$PRE/MetaXcan-master/software
export OUTPUT=$PRE/results

python $METAXCAN/SPrediXcan.py \
--gwas_file  $DATA/imputed_CARDIoGRAM_C4D_CAD_ADDITIVE.txt.gz \
--snp_column panel_variant_id --effect_allele_column effect_allele --non_effect_allele_column non_effect_allele --zscore_column zscore \
--model_db_path $MODEL/gtex_v8_mashr/mashr_Whole_Blood.db \
--covariance $MODEL/gtex_v8_mashr/mashr_Whole_Blood.txt.gz \
--keep_non_rsid --additional_output --model_db_snp_key varID \
--throw \
--output_file $OUTPUT/spredixcan/eqtl/CARDIoGRAM_C4D_CAD_ADDITIVE__PM__Whole_Blood.csv

```

## plot and interpret s-predixcan results

```{r}



```


## run multixcan (optional)

```{bash, eval=FALSE}

python $METAXCAN/SMulTiXcan.py \
--models_folder $DATA/models/eqtl/mashr \
--models_name_pattern "mashr_(.*).db" \
--snp_covariance $DATA/models/gtex_v8_expression_mashr_snp_covariance.txt.gz \
--metaxcan_folder $OUTPUT/spredixcan/eqtl/ \
--metaxcan_filter "CARDIoGRAM_C4D_CAD_ADDITIVE__PM__(.*).csv" \
--metaxcan_file_name_parse_pattern "(.*)__PM__(.*).csv" \
--gwas_file $OUTPUT/processed_summary_imputation/imputed_CARDIoGRAM_C4D_CAD_ADDITIVE.txt.gz \
--snp_column panel_variant_id --effect_allele_column effect_allele --non_effect_allele_column non_effect_allele --zscore_column zscore --keep_non_rsid --model_db_snp_key varID \
--cutoff_condition_number 30 \
--verbosity 7 \
--throw \
--output $OUTPUT/smultixcan/eqtl/CARDIoGRAM_C4D_CAD_ADDITIVE_smultixcan.txt

```

# Colocalization methods

## fine-map GWAS results - 
We will run torus due to time limitation but ideally we would like to run a method that allows multiple causal variants per locus.

```{bash, eval=FALSE}

#torus -d Height.torus.zval.gz --load_zval -dump_pip Height.gwas.pip
#gzip Height.gwas.pip

torus -d /Users/haekyungim/Box/LargeFiles/QGT-Columbia-HKI/fastenloc/data/Height.torus.zval.gz --load_zval -dump_pip /Users/haekyungim/Box/LargeFiles/QGT-Columbia-HKI/fastenloc/data/Height.gwas.pip
gzip Height.gwas.pip

```

## estimate priors
is this done internally by fastENLOC?
```{r}



```

## calculate colocalization with fastENLOC 

```{bash, eval=FALSE}
## tutorial https://github.com/xqwen/fastenloc/tree/master/tutorial

export EQTLGZ=eqtl_annotation_gzipped
export GWASGZ=gwas_data_gzipped
export TISSUE=Whole_Blood
fastenloc -eqtl EQTLGZ -gwas GWASGZ -t tissue_name #[-total_variants total_snp] [-thread n] [-prefix prefix_name] [-s shrinkage]

```

## analyze results compare with s-predixcan results

```{r}



```

# Mendelian randomization methods

## run SMR (optional)
```{bash, eval=FALSE}

```

## run  TWMR (for a locus)
```{bash, eval=FALSE}

```