Tutorial¶
Introduction¶
The Taiji software is a versatile genomics data analysis pipeline. It can be used to analyze ATAC-seq, RNA-seq, single cell ATAC-seq and Drop-seq data. However, the uniqueness and the power of Taiji really lie in its ability to integrate diverse datasets and use these information in a clever way to construct regulatory network and identify candidate driver genes.
Taiji accepts many data formats. It can start with raw data like fastq or post-processed files like bam or bed files.
How to use¶
Preparing input¶
To run the Taiji pipeline, you would need 2 configuration files.
The first configuration file is used to specify the options used by the pipeline.
Please look at this example configuration file
or Configuration for details.
The second configuration file contains the information about the input data sets.
Take a look at this example file
or Input data format.
Visualizing the workflow¶
You can use taiji view taiji.html
to generate a HTML output of the workflow.
Use a web browser to open the file and see what components are included in the
Taiji pipeline.
Running the workflow¶
To run the entire workflow, simply supply the taiji
with the first configuration file:
taiji run --config example_config.yml
.
Oftentimes you just need to run certain steps. For example, if you only want to
get some QC metrics, type taiji run --config example_config.yml --select SCATAC_QC
.
You can run multiple steps together using
taiji run --config example_config.yml --select STEP1,STEP2,STEP3
.
Parallelism and distributed computing¶
Taiji can use multiple cores. For example, to use 5 cores, simply type:
taiji run --config config.yml -n 5 +RTS -N5
.
-n 5 +RTS -N5
tells the taiji to use 5 cores/threads.
Taiji also supports distributed computing. This feature can be turned on by adding
the --cloud
flag. To use this feature, you need to have a job scheduling system like
SGE or slurm. See Advanced Usage for more details.
Auto-recovery¶
The pipeline supports auto-recovery, which means you can stop the program at any time and it will resume from the last checkpoint. The checkpoints are saved in a file called “sciflow.db”. Delete this file if you want a fresh run.
Results¶
The structure of the results in the output directory looks like this:
output_directory
+-- ATACSeq
| +-- Bam
| | +-- *.bam (raw alignment files)
| | +-- *_filt.bam (filtered alignment files)
| | +-- *_filt_dedup.bam (filtered and deduplicated alignment files)
| +-- Bed
| | +-- *.bed.gz (BED files converted from Bam files)
| | +-- *.merged.bed.gz (merged BED file from all replicates)
| +-- Download (any files that are downloaded from the web)
| +-- Peaks
| | +-- *.narrowPeak (narrow peaks called by MACS2 using loose cutoff)
| +-- TFBS
| | +-- *.bed (TF binding sites found in open chromatin)
| +-- openChromatin.bed (the union of accessible chromatin from all samples)
| +-- QC.tsv (some quality control metrics)
+-- Network (network related files)
+-- Promoters
+-- *_promoters.bed (active promoters)
+-- RNASeq
+-- *_gene_quant.tsv (gene quantification)
+-- *.bam (alignment files)
+-- expression_profile.tsv (expression profiles from all samples in one file)
+-- GeneRanks_PValues.tsv (p-values for the PageRank scores)
+-- GeneRanks.tsv (PageRank scores)