regulon-enrichment is a Python module used to predict the activity of regulatory proteins from RNAseq data.
regulon-enrichment submodules:
Load -omic datasets
Regulon utilities
regulon-enrichment requires:
- Python (>= 3.6)
- scikit-learn (>= 0.21.3)
- NumPy (>= 1.17.3)
- SciPy (>= 1.3.1)
- pandas (>= 0.25.3)
- tqdm (>= 4.38.0)
- dill (>= 0.3.1.1)
If you already have a working installation of numpy and scipy,
the easiest way to install regulon-enrichment is using ``conda`` ::
conda install -c estabroj89 regulon-enrichment
or ``pip``::
pip install regulon-enrichment==0.0.2b0
This method leverages pathway information and gene expression data to produce regulon-based protein activity scores. Our method tests for positional shifts in experimental-evidence supported networks consisting of transcription factors and their downstream signaling pathways when projected onto a rank-sorted gene-expression signature.
This regulon enrichment method utilizes pathway and molecular interactions and mechanisms available through Pathway Commons to accurately infer aberrant transcription factor activity from gene expression data.
When installing the regulon-enrichment package, the set of scripts that make up to inteface to regulon-enrichment will automatically be placed as an executables in your path, so that you can refer to these without modifying your shell environment. For example, if you install regulon-enrichment using conda, then enrich will become available on the path, and can be run as:
enrich
cohort : which cohort to use; this information will be retained in the serialized Enrichment class
expr : which tab delimited expression matrix to use shape : [n_features, n_samples], units : TPM, RPKM
out_dir : output directory - directory serialized Enrichment object and enrichment.tsv will be saved to
regulon : optional regulon containing weight interactions between regulator and
downstream members of its regulon shape : [len(Target), ['Regulator','Target','MoA','likelihood']
regulon_size : number of downstream interactions required for a given regulator in order to calculate enrichment score default=15
sec_intx : path to pre-compiled serialized secondary interaction network, default=secondary_intx_regulon.pkl
scaler_type : scaler to normalized features/samples by: standard | robust | minmax | quant, default=robust
thresh_filter : Prior to normalization remove features that have a standard deviation per feature less than {thresh_filter}, default=0.1)
To quantify the regulon enrichment for a given dataset, the command line script enrich is used.
Use --help argument to view options
enrich --help
Enrich requires three positional arguments: cohort,expr, out_dir
enrich cohort expr out_dir [regulon] [regulon_size] [sec_intx] [scaler_type] [thresh_filter]
It is recommended to run enrich with the default parameters.
enrich test tests/resources/test_expr.tsv test_enrichment_scores
The command above will generate enrichment scores for the unittest dataset test_expr.tsv and will generate and store the output under test_enrichment_scores/. In this directory test_enrichment_scores/, both the serialized Enrichment object test_enrichment.pkl and a tsv of the enrichment scores,test_regulon_enrichment.tsv will be found.
The enrichment.tsv file be shaped : [n_samples, n_regulators], where n_samples refers to the original number of samples provided in expr, while n_regulators will be determined based on the overlapping features present in the expr dataset and the regulon_size parameter.