Wrapper to make querying GEMINI databases a bit easier. Relies heavily on GEMINI https://github.com/arq5x/gemini and has a lot of expectations about fields and formats.
To install (requires conda):
conda create -n env-name python=2.7
source activate env-name
conda install pip
pip install pyyaml
conda config --add channels defaults
conda config --add channels conda-forge
conda config --add channels bioconda
conda install -c bioconda gemini
pip install git+https://github.com/burkej1/gemini-query-pythonCan be run in four modes sample, variant, table and info.
usage: gemini_wrapper [-h] {sample,variant,table,info} ...
optional arguments:
-h, --help show this help message and exit
Modes:
{sample,variant,table,info}
Mode to run in.
sample Searches for a given sample and returns a list of all
variants present in that sample
variant Searches database for given variant.
table Returns a table containing given fields and filtered
using given filtering options.
info Prints the fields present in the database
Generates a table using a given set of filters containing a given set of fields.
usage: gemini_wrapper table [-h] -i INPUT [-c PRESETS_CONFIG]
[-pf PRESETFILTER] [-ef EXTRAFILTER]
[-pF PRESETFIELDS] [-eF EXTRAFIELDS] [--nofilter]
[--flattened] [--hidesamples] [--genes GENES]
[-f FILTER] [-F FIELDS] -o OUTPUT
[--check_undrrover]
optional arguments:
-h, --help show this help message and exit
-i INPUT, --input INPUT
Input database to query.
-c PRESETS_CONFIG, --presets_config PRESETS_CONFIG
Config file containing a number of preset values with
space for user-defined presets.
-pf PRESETFILTER, --presetfilter PRESETFILTER
Preset filter options. One of: standard (Primary
annotation blocks and variants passing filters);
standard_transcripts (Standard but will prioritise a
given list of transcripts, the default); Can be
combined one of the following (separated by a comma):
lof (frameshift, stopgain, splicing variants and
variants deemed LoF by VEP); lof_pathogenic (lof and
variants classified Pathogenic by ENIGMA (using data
from BRCA exchange). E.g. -sf standard_transcripts,lof
-ef EXTRAFILTER, --extrafilter EXTRAFILTER
Additional fields to use in addition to the presets,
combined with the AND operator.
-pF PRESETFIELDS, --presetfields PRESETFIELDS
Can be 'base' (a set of basic fields), or 'explore'
which included population frequencies and various
effect prediction scores in addition to the base
fields. Can include user-defined sets of fields in the
presets.yaml file.
-eF EXTRAFIELDS, --extrafields EXTRAFIELDS
A comma separated list of fields to include in
addition to the chosen presets.
--nofilter Flag. If set will include filtered variants in the
output (DEPRECATED)
--flattened Flag. If set will output a table with one sample per
line.
--hidesamples Flag. Hide sample lists.
--genes GENES List of genes to include. If not specified will
include all
-f FILTER, --filter FILTER
Filter string in SQL WHERE structure, overwrites
presets.
-F FIELDS, --fields FIELDS
Comma separated list of fields to extract, overwrites
presets.
-o OUTPUT, --output OUTPUT
File to write sample query table to.
--check_undrrover Flag. If set the table output will include UNDR-ROVER
concordance metrics.
Searches the database for a given variant (using c. HGVS notation). The --partial option allows regex matching (not yet implemented).
Returns a list of all variants present in a given sample. Full sample ID or BS ID only can be given.
Returns a list of all fields present in the database.
There are several included filter presets:
| Name | Description |
|---|---|
| standard | All variants in vep_pick transcripts or given transcripts (by config or hard-coded). All preset filters include this unless otherwise stated. |
| lof | All Loss of Function (LoF) variants. |
| lof_pathogenic | All LoF variants and variants classified Pathogenic by BRCA exchange |
| reportable | All LoF or Pathogenic (and not Benign) by BRCA exchange variants in BRCA1, BRCA2, TP53, PALB2 and ATM:c.7271T>G. |
There are also several predefined sets of useful fields.
All reportable variants (one per line)
gemini_wrapper table -i my.db -o reportable_vars.tsv -pf reportable
All reportable variants (flattened to one sample per line and appending UNDR ROVER call information)
gemini_wrapper table -i my.db -o reportable_vars.tsv -pf reportable --flattened --check_undrrover
All variants in TP53, PALB2 and ATM with a REVEL score greater than 0.5 (-ef adds filters, -eF adds fields)
gemini_wrapper table -i my.db -o reportable_vars.tsv -pf standard \
-ef "vep_rvl_revel_score > 0.5" -eF vep_rvl_revel_score \
--genes TP53,PALB2,ATM
All variants associated with a given BSID
gemini_wrapper sample -i my.db -o test.tsv -S BS123456
All the hidesamples option can be used to suppress the sample lists if there are a lot of common variants
gemini_wrapper sample -i my.db -o test.tsv -S BS123456 --hidesamples
Search the database for a given variant and return the entry along with detailed sample information if present
gemini_wrapper variant -i my.db -o test.tsv -v "NM_000059.3:c.12345G>A"
Allow for partial matching with the --partial option.
gemini_wrapper variant -i my.db -o test.tsv -v "c.425" --partial
This can also be combined with the --genes, --extrafilter (-ef), --flattened and --check_undrrover options to further refine the query.
gemini_wrapper variant -i my.db -o test.tsv -v "c.425" --partial --genes STK11 --flattened --check_undrrover