The Trans-ABySS package has 2 main applications:

1. transabyss - RNAseq assembler at a single k-mer size
2. transabyss-merge - merge multiple assemblies from (1)

PART 1. Assembly with transabyss

• Usage
• Running transabyss
• Input reads
• Output assembly path
• k-mer size
• MPI and multi-threading
• Strand-specific assembly

PART 2. Merging assemblies with transabyss-merge

• Usage
• Running transabyss-merge
• Minimum and maximum k-mer sizes
• Output merged assembly path
• Output sequence prefixes

Please use our Google Group [email protected] for discussions and support. Existing topics can be viewed at https://groups.google.com/d/forum/trans-abyss.

You may also create issues on our GitHub repository at https://github.com/bcgsc/transabyss/issues.

transabyss --help    

transabyss --se reads.fq  

See below for assembling single-end reads and/or pair-end reads.

Supported formats are compressed (gzip/bzip) FASTQ/FASTA/SAM/QSEQ or BAM files.
Paired-end reads in FASTQ/A can be interleaved or in separate files. Paired-end reads in FASTQ/A must have the suffixes /1 or /2 in the read name.

Use options --se and --pe to specify the path(s) of single-end reads and paired-end reads, respectively. Example usages:

default: ./transabyss_2.0.0_assembly/transabyss-final.fa

Use options --name and --outdir to change the output directory and assembly name.

default: 32

Use option --kmer to adjust the k-mer size. k=32 has a good trade-off for assembling both rare and common transcripts. Using larger k-mers improve the assembly quality of common transcripts and transcripts with repetitive regions, but the assembly of rare transcripts may suffer.

default: no mpi processes; singe-threaded

Use option --threads to specify the number of threads. Use option --mpi to specify the number of MPI processes.

Only the first stage of assembly (de Bruijn graph) could benefit from MPI; all remaining stages may be multi-threaded.

Use option --SS to indicate that input reads are strand-specific. Strand-specific reads are expected to have /1 reads in reverse direction and /2 reads in forward direction, ie.

                    <-- R/1
 5' =======================> 3' transcript
    F/2 -->

Should you choose to assemble the same dataset with different settings (different k-mer sizes), you can merge the assemblies together into one FASTA file for downstream analyses (with transsabyss-analyze). When a sequence is contained in a longer sequence, the longer sequence is kept.

transabyss-merge --help

Example:

transabyss-merge a.fa b.fa --mink 32 --maxk 64

--mink and --maxk are used to specific the smallest and largest k-mer sizes in the input assemblies.

default: ./transabyss-merged.fa

Use option --out to specify the output path.

Use option --prefixes to specify the prefixes of output sequences.

Example:

transabyss-merge a.fa b.fa c.fa --mink 32 --maxk 64 --prefix k32. k48. k64.

One prefix for sequences from each input assembly FASTA file. This feature helps you keep track of the origin of each seqeunce in the merged assembly.