seqpull

Use a file of gene queries to pull similar sequences from short-read nucleotide datasets, assemble gene contigs, reverse complement, and trim sequences based on blast results.

Functions:

• Pull gene sequences from datasets using a high throughput snakemake pipeline.
• CAP3 assembly of pulled sequences (useful for short-read sequences, not long-read).
• Autoreverse complements sequences based on blast results against the top gene queries.
• Trims sequence ends based on blast results against the top gene queries.
  • note: the more robust your query sequences (taxonomy and size), the more robust trimming will be.
• Sequence size filter (default: <500bp sequences are removed).
• Sequences and output files are automatically named based on input file names and gene file names.
• Easy-to-parse directories and file names.
• Useful logs.

Use examples:

• Pull marker genes for organism identification (18S, 16S, COI).
• Find virus-related genes in organism transcriptomes.

Quickstart

Python environment

Use the provided yaml file to create the seqpull environment

git clone https://github.com/SingleEukaryote/seqpull
cd seqpull/code
conda env create -f env/seqpull.yaml

Running the seqpull pipeline

1. Put fasta files you want to pull sequences from in the data/DNAinputs directory. (must end with .fas or .fasta to be recognized)
2. Put fasta files you want to use as gene queries in the data/gene_queries directory. (must end with .fas or .fasta to be recognized)
   • Provided queries include 18S, 16S, and actin genes. Please curate your own queries depending on the target organisms and genes.
3. Move to code directory.
4. Do a dry run: snakemake -n
   • Allows you to preview every step of the pipeline and every file being generated.
5. Run bash script: bash seqpull.sh
   • Modify cores/threads first
   • Optionally, use in a queue system.
6. From the terminal you can run:
   • snakemake --cores 4 --keep-going > log.txt

Pipeline overview