@@ -18,72 +18,71 @@ A barcode translation table indicating the P1.xx used in the assay would be bene
This pipeline requires following packages to be properly installed and added to system path: GNU parallel, Bcl2fastq, fastp, zcat, STAR, bowtie2, python3, umi_tools, samtools, picard (2.14.1, newer version may result in error), R, featureCounts, read_distribution.py from RSeQC, bedtools.
The SHARE-seq-alignment scripts can be directly downloaded from the github website.\
After downloading all scripts, update the general configuration section in main script "Split_seq_example.sh":
1) myPATH # where the SHARE-seq scripts are installed. e.g. myPATH='/mnt/users/Script/share-seq-github-v1/'
2) pythohPATH # where python3 is installed e.g. pythohPATH='/usr/bin/python'
After downloading all scripts, update the general configuration section in main script "Share_seqV2_example.sh":
1) myPATH # where the SHARE-seq scripts are installed. e.g. myPATH='/mnt/users/Script/share-seq-github-v2/'
2) pythohPATH # where python3 is installed e.g. pythohPATH='/usr/bin/python/'
3) picardPATH # where picard is installed e.g. picardPATH='/mnt/bin/picard/picard.jar'
The pipeline also requres gtf files and aligner index files to be download and unziped into the right location.\
GTF files can be downloaded [here](https://drive.google.com/file/d/1HuGLf0vSHO58Ek5HibTRiwXWBn9fBMTz/view?usp=sharing).\
Bowtie2 index files (Hg19 and mm10) can be downloaded [here](https://drive.google.com/file/d/1bXIxznwirsZ6DZhqK1gw6ZKlj-UjFRhn/view?usp=sharing).\
Assuming SHARE-seq aligment scripts are installed to "/home/SHARE-seq-alignment/", the gtf files should be placed in the "/home/SHARE-seq-alignment/gtf" folder.\
The bowtie2 index files should be placed in the "/home/SHARE-seq-alignment/refGenome/bowtie2" folder.\
Three sets of index files (hg19, mm10 and hg19-mm10 combined genome) for star aligner should be prepared according to star aligner [manual](https://github.com/alexdobin/STAR), or downloded from here: [hg19](https://drive.google.com/file/d/1IXI4DP-mjh2qc-EQe1WnWJQOCVEI4KVX/view?usp=sharing), [mm10](https://drive.google.com/file/d/1n0UwzOeUbX7TIBOrcBbXjgH3i0UH-Ka5/view?usp=sharing), [combined genome](https://drive.google.com/file/d/15Z2YMUDiavYG0s9zLFAbbwqA0VhVNu-f/view?usp=sharing).\
The unziped index files should be placed in the "/home/SHARE-seq-alignment/refGenome/star/hg19", "/home/SHARE-seq-alignment/refGenome/star/mm10", and "/home/SHARE-seq-alignment/refGenome/star/both", respectively.\
Assuming SHARE-seq aligment scripts are installed to "/home/SHARE-seq-alignment/", the gtf files should be placed in the "/home/SHARE-seq-alignment/gtf/" folder.\
The bowtie2 index files should be placed in the "/home/SHARE-seq-alignment/refGenome/bowtie2/" folder.\
Four sets of index files (hg38, hg19, mm10 and hg19-mm10 combined genome) for star aligner should be prepared according to star aligner [manual](https://github.com/alexdobin/STAR), or downloded from here: [hg19](https://drive.google.com/file/d/1IXI4DP-mjh2qc-EQe1WnWJQOCVEI4KVX/view?usp=sharing), [mm10](https://drive.google.com/file/d/1n0UwzOeUbX7TIBOrcBbXjgH3i0UH-Ka5/view?usp=sharing), [combined genome](https://drive.google.com/file/d/15Z2YMUDiavYG0s9zLFAbbwqA0VhVNu-f/view?usp=sharing).\
The unziped index files should be placed in the "/home/SHARE-seq-alignment/refGenome/star/hg38", "/home/SHARE-seq-alignment/refGenome/star/hg19", "/home/SHARE-seq-alignment/refGenome/star/mm10", and "/home/SHARE-seq-alignment/refGenome/star/both", respectively.\
The index file for hg19-mm10 combined genome can be downloaded from [10x Genomics website](https://support.10xgenomics.com/single-cell-gene-expression/software/downloads/latest).
# How to run the script?
A small set of fastq files for testing are in the test_fastq_nova/ folder
Before running, three sections in the main script "Split_seq_example.sh" need to be updated for each run, inlcuding
A) paths B) sample configuration C) fastq configuration. After update all specific information in Split_seq_example.sh and config_example.ymal, run the script by ```./Split_seq_example.sh```
A) paths B) sample configuration C) fastq configuration. After update all specific information in Share_seqV2_example.sh and config.example.yaml, run the script by ```./Share_seqV2_example.sh```
## A) paths
1) rawdir=./test_fastq_nova/ # where the raw data is
2) dir=~/test/ # where output data will be stored
3) ymal=./config_example.ymal # where the ymal configuration file is
1) rawdir=./example_fastq/ # where the raw data is
2) dir=./test/ # where output data will be stored; ./example_output/ shows the output from the example fastqs
3) yaml=./config_example.ymal # where the ymal configuration file is
## B) sample configuration
1) Project=(sp.rna sp.atac.first) # use differnt name for each sample
1) Project=(BMMC.RNA BMMC.ATAC) # use differnt name for each sample
2) Type=(RNA ATAC) # ATAC or RNA
3) Genomes=(hg19 both) # both mm10 hg19 \
3) Genomes=(hg38 hg39) # both mm10 hg19 hg38 \
RawReadsPerBarcode and ReadsPerBarcode options are designed to remove barcodes with too few reads and speed up processing.
4) RawReadsPerBarcode=(10 10) # reads cutoff for the unfiltered bam file. Recommend to use 100 for full run; 10 for QC run
5) ReadsPerBarcode=(1 1) # reads cutoff for the filtered bam file. Recommend to use 100 for full run, 1 for QC run
6) keepMultiMapping=(F F) # T or F; default is F. Keep or discard multi-mapping reads
4) ReadsPerBarcode=(10 10) # reads cutoff to barcodes: 100 for full run; 10 for QC run
5) keepMultiMapping=(F F) # default F; F for species mixing or cell lines, T for low yield tissues (only keep the primarily aligned reads), doesn't matter for ATAC. Allowing multi-mapping redas will increase the percent of mito reads
1) Indexed=F # T or F; defaul is F. Indicate if the index reads are already attached to biological reads. Use F, when started with BCL file.
2) Start=Fastq_Merge # Bcl or Fastq_Merge (when fastq were generated per run) or Fastq_SplitLane (when fastq were generated per sequencing lane)
3) Runtype=full # QC or full; QC only analyze 12M reads to get a quick sense of data
4) Sequencer=Novaseq # Novaseq or Nextseq; miseq or nova-seq has the same sequencing direction, use "Novaseq" for either
1) Start=Fastq # Bcl or Fastq
2) Runtype=QC # QC or full, QC only analyze 12M reads to get a quick sense of data
3) chem=fwd # rev or fwd: speficy the chemistry used in sequencing, nova1.5 & nextseq use rev; nova1.0 uses fwd
## RNA-seq options
The pipeline also offers flexible RNA-seq specific options for advanced users.
1) removeSingelReadUMI=F # T or F; default is F. If T, UMIs with single read will be removed.
2) keepIntron=T # T or F; default is T. If F, intronic RNA reads will be discarded.
3) matchPolyT=F # T or F; default is F. If T, will try to find TTTTTT (allowing 1 mis-match) in 11-16 bp position of biological read2. If TTTTTT is not identified, read will be disgarded. Only works if Read2 is longer than 16 bp.
4) SkipPolyGumi=F # T or F; default is F, pipeline will remove polyG UMIs. If T, pipeline will keep polyG UMIs.
5) genename=gene_name # gene_name (official gene symbol) or gene_id (ensemble gene name), gene_name is default
6) refgene=gencode # gencode or genes; gencode is default; genes is UCSC refseq genes; gencode also indcludes nc-RNA
refgene=gencode # default gencode; gencode or genes; genes is UCSC genes; gencode also annotate ncRNA
mode=fast # fast or regular; default fast; fast: dedup with custom script; regular: dedup with umitools
fast mode gives more UMIs because taking genome position into account when dedup. It doesn't collapse UMIs map to different position. The lib size estimation is not accurate.
# Sample barcode table
SHARE-seq allows mutiplexing samples in one run. We use ymal file to store two levels of sample barcode information, including Round1 hybridization barcode (R1.xx), and PCR barcode (P1.xx, refers to the Ad1.xx primers used in the PCR step). See ```config_example.ymal``` as an example. This file needs to be updated for each sample and each sequencing run. When all the 96 barcodes in the Round1 plate are used during the experiment, see ```config_example2.yaml``` as an example.
SHARE-seq allows mutiplexing samples in one run. We use ymal file to store PCR barcode information (P1.xx, refers to the Ad1.xx primers used in the PCR step). See ```config_example.yaml``` as an example. This file needs to be updated for each sample and each sequencing run. When multiple sublibraries are sequenced at the same time, simply add addtional P1.xx to the yaml file. (e.g. P1.13)
```
---
Project1:
Name: sp.atac.first
Name: BMMC.RNA
Primer:
- P1.01
- P1.02
Round1:
- R1.05
- R1.13
- R1.21
-R1.29
...
- P1.12
- P1.13
Type: RNA
Project2:
Name: BMMC.ATAC
Primer:
-P1.04
Type: ATAC
```
R1.xx can be R1.01, R1.02, ..., R1.96.\
P1.xx can be P1.01, P1.02, ..., P1.96.\
The detialed information about these barcode can be found in [SHARE-seq manuscript](https://www.sciencedirect.com/science/article/pii/S0092867420312538).
@@ -96,7 +95,7 @@ This pipeline currently keeps many intermedia files. If preferred, they can be m
# Read data example
Public SHARE-seq datasets are available at [GEO](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE140203).\
A set of fastq files for species mixing experiment can be downloaded [here](https://drive.google.com/drive/folders/19HdjJuWrpRJz8OeB6YNUMSojPyTMV7OP?usp=sharing).
A set of fastq files for human bone marrow cells experiment can be downloaded [here](https://drive.google.com/drive/folders/19HdjJuWrpRJz8OeB6YNUMSojPyTMV7OP?usp=sharing).
# Cite us
For more details, please refer to [Ma et al. Chromatin Potential Identified by Shared Single-Cell Profiling of RNA and Chromatin, Cell 2020](https://www.sciencedirect.com/science/article/pii/S0092867420312538)
