# Trunk River metagenome-assembled genomes (MAGs) workflow
### Step 1: Sequence QC using PRINSEQ
### Step 2: Sequence Assembly using SPAdes
### Step 3: Contigs statistics using MetaQUAST
### Step 4: Binning using Anvi'o & Metabat
### Step 5: Bins statistics using CheckM (completeness and contamination)

# PRINSEQ
### Commands to calculate summary statistics

> prinseq-lite.pl -verbose -fastq <forward read> -fastq2 <reverse read> -stats_all > <sample_stats>.txt

> prinseq-lite.pl -verbose -fastq sample1_R1.fastq -fastq2 sample1_R2.fastq --graph_data sample1.graph.gd -qual_noscale -exact_only -no_qual_header

### Commands to eliminate sequences with less than 25 mean quality scores, replicated reads and sequencing tags
> prinseq-lite.pl -fastq *_R1.fastq -fastq2 *_R2.fastq -min_qual_mean 25 -ns_max_n 0 -derep 12345 -lc_method entropy -lc_threshold 50 -trim_qual_left 20 -trim_qual_right 20 -trim_qual_type min -trim_qual_rule lt -trim_qual_window 1 -trim_qual_step 1 -out_format 3 -out_good SK_good -out_bad SK_bad

# SPAdes
#### Step 1. Concatenate all forward and reverse reads
> cat *_good_1.fastq > concatenated1.fastq
> cat *_good_2.fastq > concatenated_2.fastq
#### Step 2. Run SPAdes
> spades.py -1 concat_1.fastq -2 concat_2.fastq -o TrunkMeta_assembly --meta

# MetaBat
#### Step 1: Build index using bowtie2
> bowtie2-build contigs.fasta spades-contigs-bowtie2-index
#### Step 2: Map reads to the contigs to generate the bam files
> bowtie2 -x spades-contigs-bowtie2-index -1 7A3_good_1.fastq -2 7A3_good_2.fastq -S sample_1.sam
#### Step 3: Convert SAM to BAM file using Samtools (do not clusterize)
> samtools view -bS sample_1.sam > sample_1.bam
#### Step 4: Sort BAM file
> samtools sort sample_1.bam -o sample.sorted_1.bam
#### Step 5: Summarize mapping files (Calculate the coverage)
> jgi_summarize_bam_contig_depths --outputDepth depth.txt --pairedContigs paired.txt *.bam
#### Step 6: Run the binnig step
##### https://bitbucket.org/berkeleylab/metabat

> metabat -i contigs.fasta -a depth.txt -m 2000 --saveCls -o output-file
> mkdir bin_folder
> mv output-file.*.fa bin_folder

# CheckM
### https://github.com/Ecogenomics/CheckM/wiki/Genome-Quality-Commands

#### Step 1: Place bins in the reference genome
> checkm tree <bin folder> <output folder>
> checkm tree Bins tree_folder -x fna
#### Step 2: Assess phylogenetic markers found in each bin
> checkm tree_qa <tree folder>
> checkm tree_qa tree_folder -o 1
#### Step 3: Infer lineage-specific marker sets for each bin
> checkm lineage_set <tree folder> <marker file>
> checkm lineage_set tree_folder marker_file
#### Step 4: List available taxonomic-specific marker sets
> checkm taxon_list
#### Step 5: Generate taxonomic-specific marker set
> checkm taxon_set <rank> <taxon> <marker file>
> checkm taxon_set domain Bacteria marker_file
#### Step 6: Identify marker genes bins
> checkm analyze <marker file> <bin folder> <output folder>
> checkm analyze marker_file bin_folder -x fa out_folder
#### Step 7: Assess bins for contamination and completeness
> checkm qa marker_file out_folder -o 1

# MetaQUAST
> metaquast.py contigs.fasta -o metaquast_output
> metaquast.py contigs.fasta -o metaquast_output --max-ref-number 0

# Anvi'o
### http://merenlab.org/2016/06/22/anvio-tutorial-v2/#creating-an-anvio-contigs-database
#### Step 1: Create Anvi'o contigs database
> anvi_database anvi-gen-contigs-database -f contigs.fasta -o contigs.db

#### Step 2: Run Hidden Markov model (archaeal and bacterial single-copy core genes search)
> anvi-run-hmms -c contigs.db

#### Step 3: Display contigs stats
> anvi-display-contigs-stats contigs.db
##### Open your browser and connect to localhost port 8080: http://localhost:8080

#### Step 4: Annotate genes with functions
> anvi-run-ncbi-cogs -c contigs2.db --num-threads 20

#### Step 5: Import taxonomy
> anvi-get-sequences-for-gene-calls -c contigs2.db -o gene-calls.fa

> centrifuge -f -x /blastdb/centrifuge/p+h+v/p+h+v gene-calls.fa -S centrifuge_hits.tsv

> anvi-import-taxonomy-for-genes -c contigs2.db -i centrifuge_report.tsv centrifuge_hits.tsv -p centrifuge

#### Step 6: Sort and index bam files
> anvi-init-bam sample_1.bam -O SAMPLE-01
> anvi-init-bam sample_2.bam -O SAMPLE-02
> anvi-init-bam sample_3.bam -O SAMPLE-03
> anvi-init-bam sample_4.bam -O SAMPLE-04
#### Step 8: Create Anvi'o profiles
##### --sample-name parameter: This name will appear in the graph. Sample names can't start with digits.
> anvi-profile -i SAMPLE-01.bam -c contigs.db --output-dir PROFILES/SAMPLE_01_Profile --sample-name A1

#### Step 7: Merge Anvi'o profiles
> anvi-merge SAMPLE-01/PROFILE.db SAMPLE-02/PROFILE.db SAMPLE-03/PROFILE.db -o SAMPLES-MERGED -c contigs.db
> anvi-merge */PROFILE.db -o SAMPLES-MERGED -c contigs.db
> anvi-merge PROFILES3/SAMPLE_01_Profile/PROFILE.db PROFILES3/SAMPLE_02_Profile/PROFILE.db PROFILES3/SAMPLE_03_Profile/PROFILE.db PROFILES3/SAMPLE_04_Profile/PROFILE.db -o SAMPLES-MERGED -c contigs.db --sample-name samples_merged

##### if you would like to skip default CONCOCT clustering, use --skip-concoct-binning flag
> anvi-merge PROFILES3/SAMPLE_01_Profile/PROFILE.db PROFILES3/SAMPLE_02_Profile/PROFILE.db PROFILES3/SAMPLE_03_Profile/PROFILE.db PROFILES3/SAMPLE_04_Profile/PROFILE.db -o SAMPLES-MERGED4 -c contigs2.db --sample-name Trunk_River --skip-concoct-binning

#### Step 8: anvi-interactive
> anvi-interactive -p SAMPLES-MERGED4/PROFILE.db -c contigs2.db --server-only -P 8080
##### Open your browser and connect to localhost port 8080: http://localhost:8080