Investigating the prevalence of Helicobacter Pylori in South Africa

# Wiki

## Data location:
The data in FASTQ format was sent via WeTransfer, downloaded and uploaded to ilifu
``` shell
/cbio/projects/032/rawdata/fastq
```

## Reference data:
Currently in project 033 on ilifu: Helicobacter pylori strain: MT5135 (RefSeq accession no. CP071982.1) (n=2)
``` shell
/cbio/projects/033/refs/Helicobacter_pylori_MT5135_refseq.fa
```

## Workflow

### 1. QC:
Raw reads were quality checked and trimmed using the kviljoen/fastq_QC pipeline which also trim adpters and filter quality based bbduk
``` shell
# For nextflow DSL1 pipeline
module load nextflow/22.10.7

nextflow run kviljoen/fastq_QC --reads '/cbio/projects/032/rawdata/fastq/*_R{1,2}_001.fastq.gz' -profile ilifu -resume --email "ephie.geza@uct.ac.za"
```
QC reports for raw reads and after trimming can be found in the
### 2. AMR profiling
We used three DBs for AMR ARGannot_r3, CARD_v3.0.8 and ResFinder:

ARGannot

## MLST
To type certain sample pair, we first downloaded the MLST scheme for **Pseudomonas aeruginosa** and renaming files by
``` shell
img="/cbio/users/katie/singularity_containers/6c884bc3ab5c-2017-12-15-c6ae6fedbccd.img"

singularity exec ${img} getmlst.py --species "Pseudomonas aeruginosa"
mv Pseudomonas_aeruginosa.fasta Pseudomonas.fasta
mv profiles_csv Pseudomonas_profiles_csv
mv alleles_fasta Pseudomonas_alleles_fasta
```
This was also done for **Klebsiella pneumoniae**, **Enterobacter cloacae**, **Escherichia coli#1**, **Escherichia coli#2**. It is important to note that **Serratia does not have MLST profile at February 2024**.

We now run MLST for each species using
``` shell
nextflow run /cbio/projects/033/uct-srst2/main.nf \
--reads '/cbio/projects/033/analysis/2024-01-11-fastq_QC/bbduk/Pseudomonas/*_{1,2}.fq' \
-profile ilifu \
--mlst_definitions /cbio/projects/033/analysis/02_MLST/profiles/Pseudomonas_profiles_csv \
--mlst_db /cbio/projects/033/analysis/02_MLST/profiles/Pseudomonas.fasta \
--mlst_delimiter "_" --outdir /cbio/projects/033/analysis/02_MLST \
-resume -dsl1
```
### MLST results
Most categorized alleles of the select group couldn't match with sufficient depthin the sequences of our short reads. Some fastq pairs had some **mismatches** represented by the number and an "*"

## Reasons why MLST may fail
- No Match Found i.e sequence data of the specified loci doesn't have a match in the MLST database (variations, mutations, or target genes not present in MLST DB)
- Low-Quality Sequences or ambiguous base calls in the sequenced loci may cause MLST assignment to fail
- Incomplete Sequencing - the seq coverage sholud be sufficient and cover all required loci
- Database Mismatch - the DB used for typing should be appropriate for the organism or strain
- Novel Sequence Type - if isolate carries a novel or uncharacterized sequence type not present in the MLST database, most common when studying less common or newly emerging strains.

## Antimicrobial Gene Detection
Mapping reds of each reference seq in fasta format throough *--gene_db* to report all genes covered beyond 80% (default is 90%)