Configuration¶
To get started, modify the example file below:
See sections below for explainations of each field.
Basic options¶
input¶
The path to a file specifying the input data sets. See Input data format for instructions for creating the input file.
input: "data/input.yml"
assembly¶
(Optional.) The name of the genome assembly. This is an optional parameter.
We use this parameter to automatically determine the values of genome, annotation
and motif_file.
If your genome assembly is not supported by the program, you will need to mannually
set the parameters for genome, annotation and motif_file.
We currently support the following assemblies: “GRCh38”, “hg38”, “GRCm38” or “mm10”.
assembly: "GRCh38"
Advanced options¶
genome¶
(Optional.) Complete genome in a SINGLE ungzipped FASTA file.
If not specified, this will be downloaded according to assembly.
genome: "/home/kai/genome/GRCh38/genome.fa"
annotation¶
(Optional.) Genome annotation in GTF format. For human and mouse, Gencode annotations are available at http://www.gencodegenes.org/. Very important: chromosome names in the annotations GTF file have to match chromosome names in the FASTA genome sequence file. For example, one can use ENSEMBL FASTA files with ENSEMBL GTF files, and UCSC FASTA files with UCSC FASTA files. However, since UCSC uses chr1, chr2,... naming convention, and ENSEMBL uses 1, 2, ... naming, the ENSEMBL and UCSC FASTA and GTF files cannot be mixed together.
If not specified, this will be downloaded according to assembly.
annotation: "/home/kai/genome/GRCh38/gencode.v25.annotation.gtf"
motif_file¶
(Optional.)
A plain text file containing PWMs in MEME format.
The naming convention for motifs is TF_NAME+OTHER_STRING, where
the TF_NAME should match the gene names in your annotation file.
You may include multipe PWMs for same TFs and use OTHER_STRING to distinguish
them. For example, SP1+ID1, SP1+ID2.
Taiji will combine sites found from different PWMs.
We provide human and mouse motifs here (downloaded from cisBP database):
Human
and mouse motif files.
When motif_file is not specified, these will be used according to
the value of the assembly field.
motif_file: "/home/kai/motif_databases/cisBP_human.meme"
seq_index¶
(Optional.)
This is the FILE containing GENOME SEQUENCE INDEX.
The program will detect whether the file exists.
If the index is not present, it will generate the index at the specified location.
If you leave this parameter unspecified,
the program will generate the index in the output_dir.
To avoid re-generate the index for every project, we recommend you to set this parameter mannually.
seq_index: "/home/kai/genome/GRCh38/GRCh38.index".
bwa_index¶
(Optional.)
This is the DIRECTORY containing BWA INDICES.
The program will detect whether the directory contain proper BWA indices.
If the indices are not present, it will generate the indices within the specified
directory. If you leave this parameter unspecified,
the program will generate the indices in the output_dir.
To avoid re-generate the indices for every project, we recommend you to set this parameter mannually.
bwa_index: "/home/kai/genome/GRCh38/BWAIndex/"
star_index¶
(Optional.)
This is the DIRECTORY containing STAR INDICES.
The program will detect whether the directory contain proper STAR indices.
If the indices are not present, it will generate the indices within the specified
directory. If you leave this parameter unspecified,
the program will generate the indices in the output_dir.
To avoid re-generate the indices for every project, we recommend you to set this parameter mannually.
star_index: "/home/kai/genome/GRCh38/STAR_index/"
rsem_index¶
(Optional.)
This is the DIRECTORY containing RSEM INDICES.
The program will detect whether the directory contain proper RSEM indices.
If the indices are not present, it will generate the indices within the specified
directory. If you leave this parameter unspecified,
the program will generate the indices in the output_dir.
To avoid re-generate the indices for every project, we recommend you to set this parameter mannually.
rsem_index: "/home/kai/genome/GRCh38/RSEM_index/"
callpeak_genome_size¶
(Optional.)
The effective genome size used for MACS2’s “-g/–gsize” parameter.
This will be automatically determined based on the assembly or genome file.
For human or mouse assembly, we set this parameter to “hs” or “mm”.
For other genome, we set this parameter to 0.9 * GENOME_SIZE.
The value of this parameter usually doesn’t make big difference.
callpeak_genome_size: "2.7e9"
tss_enrichment_cutoff¶
(Optional.) TSS enrichment cutoff for filtering cell in single cell ATAC-seq analysis.
tss_enrichment_cutoff: 7
external_network¶
(Optional.) External network file to be used in PageRank analysis.
external_network: "pathway.tsv"
Single cell analysis¶
cluster_resolution¶
cluster_resolution: 1
cluster_optimizer¶
(Optional.) Quality function used in graph clustering. Available options are RBConfiguration and CPM. RBConfiguration optimizes modularity and has resolution limit while CPM is resolution-limit free.
cluster_optimizer: CPM
scatac_fragment_cutoff¶
(Optional.) Used to remove cells that do not have enough fragments/reads.
scatac_fragment_cutoff: 1000
scrna_cell_barcode_length¶
The length of the cell barcode used in demultiplexing.
scrna_cell_barcode_length: 12
scrna_doublet_score_cutoff¶
(Optional.) Cutoff for doublet detection, a value between 0 and 1 reflecting how likely a “cell” is a doublet. (default is 0.5)
scrna_doublet_score_cutoff: 0.5
Distributed computing¶
The following settings are used in the cloud computing mode.
submit_cpu_format¶
The command line options for requesting cpu cores.
submit_cpu_format: "-l nodes=1:ppn=%d"
submit_memory_format¶
The command line options for requesting memory.
submit_memory_format: "-l mem=%dG"
resource¶
(Optional.) Specify the computational resources for each step.
resource:
SCATAC_Remove_Duplicates:
parameter: "-q home -l walltime=24:00:00"
SCATAC_Merged_Reduce_Dims:
parameter: "-q home -l walltime=24:00:00"
cpu: 4
memory: 80