Fungal_Identification

These scripts can be used to rapidly identify fungal taxonomic hits from metagenomic data.

What is included:

• Fungal_combo.R: R script used to combine all fungal hits into one table called Fungal_total.csv

• fungal_db_download.sh: Bash script used to download PlusPF Refseq index for kraken2 analysis

• fungal_lookup.sh: Bash script used to run the kraken2 analysis and identify fungal hits in the samples

• fung.list.txt: List of fungal genera that is used for the fungal identification. This can be changed as the database is updated

• kracken2.yml: Yml file that can be used to create a conda environment called kracken2.

Step 1: Change the working directory in fungal_lookup.sh and fungal_db_download.sh

If this is being run on a PBS or SLURM server you need to change line 11 of fungal_lookup.sh and line 11 of fungal_db_download.sh to the path of where these scripts are being run. If this is the case use the qsub options below

Step 2: Download kraken database

The first script to use is fungal_db_download.sh. This will download the PlusPF Refseq index. This index has the standard bacterial Refseq database with fungi and protozoa references added. Additionally, this script will create a conda environment called kracken2 from the included .yml file and the fastq directory for your fastq files.

bash ./fungal_db_download.sh

or

qsub fungal_db_download.sh

Step 3: Move Fastq data to the newly created fastq directory.

Move your paired-end Illumina fastq data to the included fastq directory. It is important to note that this data must be in _R1.fastq/_R2.fastq format. So for example, if you were working on a human sample with the ID SRA56892, the paired-end fastq data would be SRA56892_R1.fastq and SRA56892_R2.fastq. Publically available data should already be in this format.

mv Path/To/Your/Fastq/Data/*.fastq ./fastq

Step 4: Run fungal_lookup.sh

This script will run kraken2 on all paired-end fastq files in the fastq directory. This will create a directory called c_out which will include all of the classified reads in paired read format. Additionally, all the taxonomy results will be sent to a second created directory called out. Next, the script will pull out all the fungal hits and count how many times each one appears in each sample. These results will be in a created directory called fungal_hits. Finally, this script will run an R script called Fungal_combo.R. The R script will take all of the files in the fungal_hits directory and combine them into one large table called Fungal_total.csv. This file can be opened in Excel and has the tabulated fungal hits, the total reads, and the percent of fungal reads for all files included.

bash ./fungal_lookup.sh

or

qsub fungal_lookup.sh