Analysis for Manuscript

[rx32940]

• kneaddata: sequence quality check, remove contamination and host
• kraken2 mini database (most up_to_date version (minikraken_8GB_20200312) includes human genome)
  • absolute bar plot
• kraken2 w/standard database
  • absolute bar plot
• kraken2 w/custom database (?)
  • phylum
  • genus
  • species
• CLARK
  • phylum
  • genus
  • species
• CLARK-S
  • phylum
  • genus
  • species
• PCA (with results from custom databases)
• relative abundance

[rx32940]

new kraken2 analysis outputs in:

/scratch/rx32940/kraken2_052020

kneaddata

• script in for kneaddate found in kraken2_pipeline.sh
• reference for R.rattus (GCF_011064425.1) and R.norvegicus (GCF_000001895.5) from refseq
  • reference database (bowtie2-build):

/scratch/rx32940/kraken2_052020/kneaddata/ref_db

• pairend output sequences after trimming and host cleaning (base on each sample's specific host)

/scratch/rx32940/kraken2_052020/kneaddata/hostclean_seq

kraken2 databases

kraken2 database found in:

/scratch/rx32940/kraken2_052020/kraken2/kraken2_db

• build standard database
  • Do I need to do custom database?
• download minikraken_8GB_202003.tgz pre-built database from JHU ftp link (wget)
  • human genome included
  • most up_to_date database

kraken2 databases

Kraken2 results found in:

/scratch/rx32940/kraken2_052020/kraken2/(mini_output/standard_output)

• download kreport for absolute reads analysis
• local absolute reads data store in:

/Users/rx32940/Dropbox/5.Rachel-projects/Metagenomic_Analysis/final_analysis/kraken2/(minikraken/standard)/absolute

---

Rmarkdown scrip for barplot visualization with minikraken db

Rmarkdown scrip for barplot visualization with standard kraken2 db

Rmarkdown scrip for barplot visualization with custom kraken2 db

[rx32940]

new clark analysis outputs in:

/project/lslab/Rachel/Metagenomics/clark_0613

• script for CLARK analysis pipeline
• Used hostcleaned sequences by kneaddata processed by kraken2_pipeline.sh
• used tool fastq-pair to match the forward and reverse fastq into the same order

Final input for CLARK analysis:

clark_0613/hostclean_seq/{$SAMPLE_ID}_1_kneaddata_paired_{1/2}.fastq.paired.fq

CLARK database

• CLARK standard database matching KRAKEN2 standard database (with UniVec_Core.fasta and no Rattus reference sequences)
• CLARK Custom database matching KRAKEN2 custom database built in the last thread
• CLARK-S Spaced database built based on the CLARK custom database
  Dir for all three databases: clark_0613/database/standard
  - CLARK analysis will produce lineage information for each taxon, thus can only set-target for species classification to save time

CLARK output

standard database: clark_0613/output_genus(phylum)
custom database: clark_0613/output_species_rat
spaced database: output_species_rat_spaced

script for abundance visualization

Rmarkdown scrip for barplot visualization with standard db
Rmarkdown scrip for barplot visualization with custom db
Rmarkdown scrip for barplot visualization with spaced db