jpayne@17: # SeqSero2
jpayne@17: Salmonella serotype prediction from genome sequencing data.
jpayne@17: 
jpayne@17: Online version: http://www.denglab.info/SeqSero2
jpayne@1: 
jpayne@1: # Introduction 
jpayne@7: SeqSero2 is a pipeline for Salmonella serotype prediction from raw sequencing reads or genome assemblies
jpayne@1: 
jpayne@1: # Dependencies 
jpayne@17: SeqSero2 has three workflows:
jpayne@1: 
jpayne@7: (A) Allele micro-assembly (default). This workflow takes raw reads as input and performs targeted assembly of serotype determinant alleles. Assembled alleles are used to predict serotype and flag potential inter-serotype contamination in sequencing data (i.e., presence of reads from multiple serotypes due to, for example, cross or carryover contamination during sequencing). 
jpayne@1: 
jpayne@7: Allele micro-assembly workflow depends on:
jpayne@1: 
jpayne@7: 1. Python 3;
jpayne@7: 
jpayne@17: 2. Biopython 1.73;
jpayne@7: 
jpayne@17: 3. [Burrows-Wheeler Aligner v0.7.12](http://sourceforge.net/projects/bio-bwa/files/);
jpayne@7: 
jpayne@17: 4. [Samtools v1.8](http://sourceforge.net/projects/samtools/files/samtools/);
jpayne@7: 
jpayne@17: 5. [NCBI BLAST v2.2.28+](https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastDocs&DOC_TYPE=Download);
jpayne@7: 
jpayne@17: 6. [SRA Toolkit v2.8.0](http://www.ncbi.nlm.nih.gov/Traces/sra/sra.cgi?cmd=show&f=software&m=software&s=software);
jpayne@7: 
jpayne@17: 7. [SPAdes v3.15.5](http://bioinf.spbau.ru/spades);
jpayne@7: 
jpayne@17: 8. [Bedtools v2.17.0](http://bedtools.readthedocs.io/en/latest/);
jpayne@17: 
jpayne@17: 9. [SalmID v0.11](https://github.com/hcdenbakker/SalmID).
jpayne@7: 
jpayne@7: 
jpayne@7: (B) Raw reads k-mer. This workflow takes raw reads as input and performs rapid serotype prediction based on unique k-mers of serotype determinants. 
jpayne@7: 
jpayne@7: Raw reads k-mer workflow (originally SeqSeroK) depends on:
jpayne@1: 
jpayne@1: 1. Python 3;
jpayne@1: 2. [SRA Toolkit](http://www.ncbi.nlm.nih.gov/Traces/sra/sra.cgi?cmd=show&f=software&m=software&s=software) (optional, just used to fastq-dump sra files);
jpayne@1: 
jpayne@1: 
jpayne@7: (C) Genome assembly k-mer. This workflow takes genome assemblies as input and the rest of the workflow largely overlaps with the raw reads k-mer workflow
jpayne@1: 
jpayne@17: # Installation
jpayne@17: ### Conda
jpayne@17: To install the latest SeqSero2 Conda package (recommended):  
jpayne@17: ```
jpayne@17: conda install -c bioconda seqsero2=1.3.1
jpayne@17: ```
jpayne@17: ### Git
jpayne@17: To install the SeqSero2 git repository locally:
jpayne@17: ```
jpayne@17: git clone https://github.com/denglab/SeqSero2.git
jpayne@17: cd SeqSero2
jpayne@17: python3 -m pip install --user .
jpayne@17: ```
jpayne@17: ### Other options
jpayne@17: Third party SeqSero2 installations (may not be the latest version of SeqSero2): \
jpayne@17: https://github.com/B-UMMI/docker-images/tree/master/seqsero2 \
jpayne@17: https://github.com/denglab/SeqSero2/issues/13
jpayne@17: 
jpayne@1: 
jpayne@1: # Executing the code 
jpayne@1: Make sure all SeqSero2 and its dependency executables are added to your path (e.g. to ~/.bashrc). Then type SeqSero2_package.py to get detailed instructions.
jpayne@1: 
jpayne@1:     Usage: SeqSero2_package.py 
jpayne@1: 
jpayne@7:     -m <string> (which workflow to apply, 'a'(raw reads allele micro-assembly), 'k'(raw reads and genome assembly k-mer), default=a)
jpayne@1: 
jpayne@17:     -t <string> (input data type, '1' for interleaved paired-end reads, '2' for separated paired-end reads, '3' for single reads, '4' for genome assembly, '5' for nanopore reads (fasta/fastq))
jpayne@1: 
jpayne@1:     -i <file> (/path/to/input/file)
jpayne@1: 
jpayne@1:     -p <int> (number of threads for allele mode, if p >4, only 4 threads will be used for assembly since the amount of extracted reads is small, default=1) 
jpayne@1: 
jpayne@1:     -b <string> (algorithms for bwa mapping for allele mode; 'mem' for mem, 'sam' for samse/sampe; default=mem; optional; for now we only optimized for default "mem" mode)
jpayne@1:  
jpayne@1:     -d <string> (output directory name, if not set, the output directory would be 'SeqSero_result_'+time stamp+one random number)
jpayne@1: 	
jpayne@17:     -c <flag> (if '-c' was flagged, SeqSero2 will only output serotype prediction without the directory containing log files)
jpayne@17:     
jpayne@17:     -n <string> (optional, to specify a sample name in the report output)
jpayne@17:     
jpayne@17:     -s <flag> (if '-s' was flagged, SeqSero2 will not output header in SeqSero_result.tsv)
jpayne@17: 		    
jpayne@17:     --check <flag> (use '--check' flag to check the required dependencies)
jpayne@17:     
jpayne@17:     -v, --version (show program's version number and exit)
jpayne@1: 	
jpayne@1: 
jpayne@1: # Examples
jpayne@7: Allele mode:
jpayne@7: 
jpayne@7:     # Allele workflow ("-m a", default), for separated paired-end raw reads ("-t 2"), use 10 threads in mapping and assembly ("-p 10")
jpayne@7:     SeqSero2_package.py -p 10 -t 2 -i R1.fastq.gz R2.fastq.gz
jpayne@7: 	
jpayne@1: K-mer mode:
jpayne@1: 
jpayne@7:     # Raw reads k-mer ("-m k"), for separated paired-end raw reads ("-t 2")
jpayne@7:     SeqSero2_package.py -m k -t 2 -i R1.fastq.gz R2.fastq.gz
jpayne@1: 
jpayne@7:     # Genome assembly k-mer ("-t 4", genome assemblies only predicted by the k-mer workflow, "-m k")
jpayne@7:     SeqSero2_package.py -m k -t 4 -i assembly.fasta
jpayne@1: 	
jpayne@1: # Output 
jpayne@17: Upon executing the command, a directory named 'SeqSero_result_Time_your_run' will be created. Your result will be stored in 'SeqSero_result.txt' in that directory. And the assembled alleles can also be found in the directory if using "-m a" (allele mode).
jpayne@1: 
jpayne@1: 
jpayne@1: # Citation
jpayne@17: Zhang S, Den-Bakker HC, Li S, Dinsmore BA, Lane C, Lauer AC, Fields PI, Deng X. 
jpayne@17: SeqSero2: rapid and improved Salmonella serotype determination using whole genome sequencing data.
jpayne@17: **Appl Environ Microbiology. 2019 Sep; 85(23):e01746-19.** [PMID: 31540993](https://aem.asm.org/content/early/2019/09/17/AEM.01746-19.long) 
jpayne@17: 
jpayne@1: Zhang S, Yin Y, Jones MB, Zhang Z, Deatherage Kaiser BL, Dinsmore BA, Fitzgerald C, Fields PI, Deng X.  
jpayne@1: Salmonella serotype determination utilizing high-throughput genome sequencing data.  
jpayne@17: **J Clin Microbiol. 2015 May;53(5):1685-92.** [PMID: 25762776](http://jcm.asm.org/content/early/2015/03/05/JCM.00323-15)