Skip to content

Latest commit

 

History

History
37 lines (28 loc) · 3.04 KB

FACS.md

File metadata and controls

37 lines (28 loc) · 3.04 KB
Raw

ISSAAC-seq by FACS

This section contains information about ISSAAC-seq using a plate-based workflow. FACS is commonly used to sort single nuclei into wells, but handpick should also work.

About the libraries and files

Read Description
ATAC Read 1 (R1) gDNA read
Read 2 (R2) gDNA read
Index 1 (I1) i7 index
Index 2 (I2) i5 index
---------- -------------- --------------------------------------------------------------------------------------------
RNA Read 1 (R1) The first 10 bp are UMIs, the rest are ignored as they are mostly dT
Read 2 (R2) cDNA sequence, might have some adaptor contamination depending on how long you sequence it
Index 1 (I1) i7 index
Index 2 (I2) i5 index

Check this page to see how a step-by-step guide of how the libraries are generated. In this workflow, single nuclei are sorted into invidividual wells containing index primers. The library preparation is done individually. In this case, the well barcode is the cell barcode. The combination of Index 1 (i7) and Index 2 (i5) defines a cell. There are two common scenarios:

  1. You equence your libraries from a core facility. In this case, you probably need to provide the index information Index 1 (i7) + Index 2 (i5), and the core will demultiplex for you. In this case, you have two fastq files for each cell per modality: R1.fastq.gz and R2.fastq.gz.

  2. You sequence by yourself, and run bcl2fastq from Illumina to generate fastq files. You can certainly put the Index 1 (i7) + Index 2 (i5) information in the SampleSheet.csv, and each cell will get demultiplexed by bcl2fastq. A simpler way is simply put NNNNNNNN + NNNNNNNN in the SampleSheet.csv. Then run bcl2fastq like this:

bcl2fastq -r 4 -p 4 -w 4 --create-fastq-for-index-reads --no-lane-splitting -o /path/to/output/dir

Due to a known bug in bcl2fastq, the program look for the literal N in the index. Therefore, you will get the following four files:

Undetermined_S0_L001_R1_001.fastq.gz
Undetermined_S0_L001_R2_001.fastq.gz
Undetermined_S0_L001_I1_001.fastq.gz
Undetermined_S0_L001_I2_001.fastq.gz

You can start from there, and use the Snakefile to process them.