Computational Biology Tools

@crankysparrow & @bbitarello

Clone this repo

Either into a directory in your computer by running

git clone https://github.com/crankysparrow/computationalbio.git

or by clicking 'open in github desktop'(above, on the right)

An API query tool for gnomAD

Requirements:

-Node.js

-npm

You can follow this guide and simply run:

npm install

Now that the setup is complete, you can use the tool.

First, you need to provide an input file, containing SNP IDs, one per line, like this one. Put it in the same directory where you cloned the repo and use this exact name ('rsIDS.txt') for that file.

Next, convert that into a json format file like this one by running:

node scripts/readIDs.js --input input/rsIDs_whi.txt --output input/rsIDs_whi.json

Note: replace input/rsIDs_whi.txt by path/your_file_one_SNP_id_per_line

Now you can run the query and save the output into a file like this one:

node scripts/gnomad_query.js --input input/rsIDs_jhs.json --pop 'NFE' --db 'GNOMAD'

Note: default values are: NFE (pop), GNOMAD (db). There is no default input file so you must provide at least that. The output will contain 'GNOMAD-GENOMES' and 'GNOMAD-EXOMES' (in the absence of the former) allelic frequencies for each SNP in specified population.

#A fragment of the output:

[
  {
    "study": "GNOMAD_GENOMES",
    "population": "NFE",
    "refAllele": "T",
    "altAllele": "C",
    "refAlleleFreq": 0.6836476,
    "altAlleleFreq": 0.31635246,
    "refHomGenotypeFreq": 0.4681107,
    "hetGenotypeFreq": 0.43107367,
    "altHomGenotypeFreq": 0.10081562,
    "id": "rs2465136"
  },
  {

UPDATE (Aug 2nd): We have now included all alterante allele frequencies in the output, i.e., you get query results for all alternative alleles for SNPs that are poly-allelic.

LiftOver PLINK 1 files (easy)

#run 
scripts/LiftOverPlink.sh -c tmp/hg19ToHg38.over.chain.gz -i to_delete/UKB_EUR_hg38_filt -o to_delete/UKB_EUR_hg19_filt -d tmp-dir/ -m scripts/PlinkLiftOver.py -a scripts/rmBadLifts.py

Stay tuned: we might include other query options in the future. Feedback welcome!