Description
Perform TF-IDF normalization of the data (typically, ATAC).
TF-IDF stands for "term frequency - inverse document frequency". It is a technique from natural language processing analysis.
In the context of ATAC data, "terms" are the features (genes) and "documents" are the observations (cells).
TF-IDF normalization is applied to single-cell ATAC-seq data to highlight the importance of specific genomic regions (typically peaks)
across different cells while down-weighting regions that are commonly accessible across many cells.
Run the following command to execute this component with Nextflow:
cat > params.yaml <<'EOM'
modality: [ "atac" ]
output: "$id.$key.output"
output_layer: [ "tfidf" ]
scale_factor: [ 10000 ]
log_idf: [ true ]
log_tf: [ true ]
log_tfidf: [ false ]
id: "run"
publish_dir: "output/"
EOM
nextflow run https://packages.viash-hub.com/vsh/openpipeline.git \
-revision v4.0.2 \
-main-script target/nextflow/transform/tfidf/main.nf \
-params-file params.yaml Name | Type & Properties |
|---|---|
--input -i | file required |
--modality | string |
--input_layer | string |
--output -o | file required output |
--output_layer | string |
--scale_factor | integer |
--log_idf | boolean |
--log_tf | boolean |
--log_tfidf | boolean |
--output_compression | string |