Quantaf

Introduction

Many single-cell and single-nucleus datasets are publicly available. Some of these datasets are used over and over again to develop new methods, demonstrate new tools, to use in tutorials, etc. For example, 10x Genomics provided various publicly available datasets on their website for free downloading (https://www.10xgenomics.com/resources/datasets), and many of these are commonly used in single-cell tutorials and vignettes.

To stop re-inventing the wheel, here we introduce an alevin-fry workflow written in Nextflow that can be used to quantify arbitrary number of single-cell RNA-sequencing projects (https://github.com/COMBINE-lab/10x-requant) in one command with some required spreadsheets under the input_files directory as the input, which are:

sample_sheet.tsv: This spreadsheet records the detailed information of the datasets one would like to process. Please refer to the provided sheet for examples.
ref_sheet.tsv: This spreadsheet include the reference used to make the alevin-fry splici reference upon. Currently the provided reference must be the pre-build Cell Ranger references, for example, human2020A and mm10-2020A. All the references specified in the reference column of the sample_sheet.tsv must be included in the ref_sheet.tsv, otherwise the workflow will have no idea what reference should be used for mapping reads against.
pl_sheet.tsv: This spreadsheet records the permit lists that will be used for deduplicating cellular barcode in alevin-fry, for example, the permit list used for 10x Chromium V2 and V3 chemistry. All the chemistry specified in the chemistry column of the sample_sheet.tsv must be included in the pl_sheet.tsv.

With the three required spreadsheets, this workflow will download the references, make the splici references and indices, and run the alevin-fry pipeline for each dataset in the sample_sheet.tsv.

The outputs of this workflow is under the nf_pipeline/alevin_fry folder, which includes the quantification result of all the processed datasets. The quantification folder is named by the MD5sum of the dataset for simplicity. One can find the name and URL of the dataset in the dataset_description.txt file in each quantification result folder.

processed 10x datasets:

Using this workflow we have collected and processed some datasets from 10x website. Here we provide the link of the quantification result generated by alevin-fry. For more information, please check the webpage of quantaf:

About

BSD 3-Clause "New" or "Revised" License

Languages

Language:Nextflow 100.0%