Andrew Severin

Andrew Severin

His PhD was in Biophysics/NMR spectroscopy. He did a Bioinformatics Postdoc in Soybean genetics and now runs the Genome Informatics Facility at Iowa State University. He is passionate about evolution and the science behind the genome. There is so much we don't know about how the elements in a genome interact to create the fine balance of gene expression, modification and 3D structure that create the dynamic range of phenotypes we observe. As sequencing technology continues to improve and the cost continues to decrease, we will be able to ask more complex questions that increase our understanding via comparative and translational genomics.

Background

How to run SPAdes

  • SPAdes parameter considerations

    1
2
3
4
5
6
7

    --careful		tries to reduce number of mismatches and short indels
--tmp-dir	<dirname>	directory for temporary files
-t/--threads	<int>		number of threads
-m/--memory	<int>		RAM limit for SPAdes in Gb (terminates if exceeded) [default 250 (Gb)]
--sanger	<filename>	file with Sanger reads
--pacbio	<filename>	file with PacBio reads
--nanopore	<filename>	file with Nanopore reads

  • Example SLURM Job

    1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

    #!/bin/bash
#SBATCH -J spades4.e%j
#SBATCH -o spades4.o%j
#SBATCH --mem 1000GB
#SBATCH -p LM
#SBATCH -N 1
#SBATCH -n 40
#SBATCH -t 48:00:00

source /usr/share/Modules/init/bash
module load spades/3.11.1

spades.py \
-k 21,33,55,77,99,127 \
-t 16 -m 1000 \
--pacbio /pylon5/mc48o5p/severin/spadesTutorial/00_rawdata/SRR2093876_subreads.fastq.gz \
--pe1-1 /pylon5/mc48o5p/severin/spadesTutorial/00_rawdata/SRR2093871_1.fastq.gz  \
--pe1-2 /pylon5/mc48o5p/severin/spadesTutorial/00_rawdata/SRR2093871_2.fastq.gz  \
--mp1-1 /pylon5/mc48o5p/severin/spadesTutorial/00_rawdata/SRR2093872_1.fastq.gz  \
--mp1-2 /pylon5/mc48o5p/severin/spadesTutorial/00_rawdata/SRR2093872_2.fastq.gz  \
--careful \
-o BT

Expected files generated during assembly

SPAdes files output from assembly
K21 K33 K55
K77 K99 assembly_graph.fastg
assembly_graph_with_scaffolds.gfa before_rr.fasta contigs.fasta
contigs.paths corrected dataset.info
input_dataset.yaml misc mismatch_corrector
params.txt scaffolds.fasta scaffolds.paths
spades.log tmp warnings.log

SLURM standard output

  • An example of the output log can be found here: spades.log

Errors

  • Not enough memory

    These two errors are extremely unhelpful for diagnosing the problem. If you get them, try a computer with much higher memory or running spades single threaded (-t 1). Also, I have seen multiple err code message numbers while attempting multithreaded spades on a low memory machine (<256Gb RAM)

    1

    libgomp: Thread creation failed: Resource temporarily unavailable

    1

    finished abnormally, err code: 1

  • Remove spaces in fastq files

    If you still are running into errors, you may also consider removing spaces in fastq sequence headers as follows.

    1
2
3
4
5
6
7
8
9
10
11
12

    gunzip SRR2093871_1.fastq.gz &
gunzip SRR2093871_2.fastq.gz &
gunzip SRR2093872_1.fastq.gz &
gunzip SRR2093872_2.fastq.gz &
perl -i -pe 's/ /_/g' SRR2093871_1.fastq &
perl -i -pe 's/ /_/g' SRR2093871_2.fastq &
perl -i -pe 's/ /_/g' SRR2093872_2.fastq &
perl -i -pe 's/ /_/g' SRR2093872_1.fastq &
gzip SRR2093871_1.fastq &
gzip SRR2093871_2.fastq &
gzip SRR2093872_1.fastq &
gzip SRR2093872_2.fastq &

Further Reading

Back to the Assembly and Annotation Index page