Assembly Nodes
Assembly nodes reconstruct genomes or transcriptomes de novo from reads. They are among the most memory-intensive operations in bioinformatics — large eukaryotic assemblies can need hundreds of gigabytes of RAM.
SPAdes
Versatile de Bruijn graph assembler for bacterial and small eukaryotic genomes (also single-cell and metagenomic modes).
Inputs: reads: FASTQ
Outputs: contigs: FASTA, scaffolds: FASTA
| Parameter | Type | Default | Description |
|---|---|---|---|
mode | String | isolate | isolate / sc / meta |
kmers | String | auto | Comma-separated k-mer sizes |
threads | Int | 8 | Worker threads |
Trinity
De novo transcriptome assembly from RNA-seq reads.
Inputs: reads: FASTQ
Outputs: transcripts: FASTA
| Parameter | Type | Default | Description |
|---|---|---|---|
maxMemory | String | 64G | Max memory for the Jellyfish step |
minContigLen | Int | 200 | Minimum reported transcript length |
QUAST (assembly QC)
Evaluates assembly quality (N50, misassemblies, completeness) and produces a report.
Inputs: assembly: FASTA, reference: FASTA (optional)
Outputs: report: Report
Memory guidance
| Assembly target | Suggested tier |
|---|---|
| Bacterial genome (SPAdes) | Medium (64 GB) |
| Plant/animal transcriptome (Trinity) | Large (128 GB)+ |
| Large eukaryotic genome | XLarge (256 GB) / Extreme (1 TB) |
Assemblies are the prime use case for checkpointing: a multi-day run that hits a spot-instance eviction resumes from its last checkpoint rather than restarting.