.. _index: .. image:: lOGO_3_transparente.png :width: 250 :align: left :alt: Laboratory logo ---- NGS-preprocess ************** `NGS-preprocess `_ is a pipeline developed with `Nextflow `_ and `Docker `_. It was designed to provide an easy-to-use framework for preprocessing sequencing reads from Illumina, Pacbio and Oxford Nanopore platforms. It wraps up the following tools: .. list-table:: :widths: 10 60 40 :header-rows: 1 * - Software - Analysis step - Source * - sra-tools & entrez-direct - Interaction with SRA database for fetching fastqs and metadata - https://anaconda.org/bioconda/entrez-direct ; https://github.com/ncbi/sra-tools * - Fastp - tool designed to provide fast all-in-one preprocessing for FastQ files - https://github.com/OpenGene/fastp * - Porechop - ONT reads trimming and demultiplexing - https://github.com/rrwick/Porechop * - pycoQC - ONT reads QC - https://github.com/tleonardi/pycoQC * - NanoPack - Long reads QC and filter - https://github.com/wdecoster/nanopack * - bax2bam - Convert PacBio bax files to bam - https://github.com/PacificBiosciences/bax2bam * - bam2fastx - Extract reads from PacBio bam files - https://github.com/PacificBiosciences/bam2fastx * - lima - PacBio reads demultiplexing - https://github.com/PacificBiosciences/barcoding * - pacbio ccs - Generate PacBio Highly Accurate Single-Molecule Consensus Reads - https://ccs.how/ .. toctree:: :hidden: installation profiles quickstart manual config examples Support Contact *************** Feel free to contact me at almeidafmarques@gmail.com