eDNA Reference Database Creation Pipeline

SOP ID: EDNA-005 Version: 2.0 Date: October 2025 Author: AGRP CRABS Version: 1.12.0+

---

Overview

This pipeline creates curated multi-marker reference databases for eDNA metabarcoding analysis of South African marine species in the Southern Atlantic and Indian Ocean. It uses CRABS (Creating Reference databases for Amplicon-Based Sequencing) to integrate sequences from four databases:

Database	Sequences	Strength
BOLD	COI-5P barcodes	Gold standard for COI; curated, standardised metadata
NCBI	12S, 16S, COI	Broadest taxonomic coverage including regional species
EMBL	Vertebrate mitochondrial	European sequence data
MitoFish	Fish mitochondrial genomes	Complete fish mitogenomes

Target Taxa

• Fishes: Actinopterygii, Chondrichthyes, Sarcopterygii (inc. Latimeria chalumnae), Myxini, Petromyzontida
• Marine Mammals: Cetacea, Pinnipedia (inc. Cape fur seal Arctocephalus pusillus)

Target Markers

Marker	Target	Amplicon	Purpose
MiFish-U (12S rRNA)	Fish, Cetaceans	160–180 bp	Universal fish detection
MarVer1 (12S rRNA)	Cetaceans	~202 bp	Marine mammal specific
MarVer3 (16S rRNA)	Pinnipeds, Sirenians	~245 bp	Marine mammal specific
STAT (16S rRNA)	Cetaceans, marine mammals	70–80 bp	Marine mammal specific
Leray-XT (COI)	Fish, Cetaceans, Invertebrates	~313 bp	Species-level ID

Primer Sequences

Marker	Direction	Sequence
MiFish-U	Forward	GTCGGTAAAACTCGTGCCAGC
MiFish-U	Reverse	CATAGTGGGGTATCTAATCCCAGTTTG
MarVer1	Forward	CGTGCCAGCCACCGCG
MarVer1	Reverse	GGGTATCTAATCCYAGTTTG
MarVer3	Forward	AGACGAGAAGACCCTRTG
MarVer3	Reverse	GGATTGCGCTGTTATCCC
STAT	Forward	TTAGACACTTTGGGAGGCTG
STAT	Reverse	CTGTGTAGCCGAAGGTAGAC
Leray-XT	Forward	GGWACWRGWTGRACWITITAYCCYCC
Leray-XT	Reverse	TAIACYTCIGGRTGICCRAARAAYCA

---

Workflow Summary

Phase 1: Setup & Taxonomy Files
Phase 2: Download sequences (BOLD, NCBI, EMBL, MitoFish)
Phase 3: Import & standardise all sequences
    ↓
Workflow A: MiFish-U (12S)   → Merge → PCR → Filter → Derep → Subset → Export
Workflow B: MarVer (16S/12S) → Merge → PCR → Filter → Derep → Subset → Export
Workflow C: Leray-XT (COI)   → Merge → PCR → Filter → Derep → Subset → Export
    ↓
Phase 7: Create BLAST databases
Phase 8: Visualisation & QC
Phase 9: Downstream analysis (BLAST / DADA2 / QIIME2)

---

Prerequisites

Software

conda --version       # Conda must be installed
conda activate crabs
crabs --version       # Requires CRABS 1.12.0+

Species List (merged_sa_species.txt)

The pipeline filters databases to South African marine species. This file must be prepared before starting:

# Convert FishBase download to pipeline format
bash prepare_species_lists.sh --convert-fish

# Merge fish and cetacean lists
bash prepare_species_lists.sh --merge

# Verify output (~2,087 species)
wc -l species_lists/merged_sa_species.txt
head -5 species_lists/merged_sa_species.txt

Species are downloaded from FishBase and SeaLifeBase. See the prepare_species_lists.sh script for full details.

---

Phase 1: Environment Setup & Taxonomy Files

1.1 Activate CRABS Environment

conda activate crabs
crabs --version

1.2 Create Project Directory Structure

mkdir -p raw_data processed_data reference_dbs taxonomy results
cd /home/evilliers/work/edna_method/edna_sa_project

1.3 Download NCBI Taxonomy Files

One-time setup — ~13 GB, takes 30–60 minutes

These files are required for all import operations. Download them once and reuse.

cd taxonomy

# NCBI taxonomy database
wget ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/taxdump.tar.gz
tar -xzf taxdump.tar.gz

# Accession-to-taxonomy mapping
wget ftp://ftp.ncbi.nlm.nih.gov/pub/taxonomy/accession2taxid/nucl_gb.accession2taxid.gz
gunzip nucl_gb.accession2taxid.gz

cd ..

Expected files in taxonomy/:

File	Size
names.dmp	~272 MB
nodes.dmp	~205 MB
nucl_gb.accession2taxid	~13 GB

---

Phase 2: Download Sequence Data

2.1 Download from BOLD (COI-5P only)

# Ray-finned fishes
crabs --download-bold --taxon 'Actinopterygii' --marker 'COI-5P' --version-v3 --output raw_data/bold_actinopterygii_coi.fasta

# Cartilaginous fishes
crabs --download-bold --taxon 'Chondrichthyes' --marker 'COI-5P' --version-v3 --output raw_data/bold_chondrichthyes_coi.fasta

# Coelacanths
crabs --download-bold --taxon 'Sarcopterygii' --marker 'COI-5P' --version-v3 --output raw_data/bold_sarcopterygii_coi.fasta

# Hagfishes
crabs --download-bold --taxon 'Myxini' --marker 'COI-5P' --version-v3 --output raw_data/bold_myxini_coi.fasta

# Lampreys
crabs --download-bold --taxon 'Petromyzonti' --marker 'COI-5P' --version-v3 --output raw_data/bold_lampreys_coi.fasta

# Cetaceans
crabs --download-bold --taxon 'Cetacea' --marker 'COI-5P' --version-v3 --output raw_data/bold_cetacea_coi.fasta

# All Chordata (includes Pinnipeds/seals)
crabs --download-bold --taxon 'Chordata' --marker 'COI-5P' --version-v3 --output raw_data/bold_chordata_coi.fasta

Manual download fallback

If the automated download fails, download manually from boldsystems.org → Public Data Portal → Taxonomy Browser. Filter by COI-5P and download as FASTA.

2.2 Download from NCBI

Note

Replace your.email@example.com with your actual email address. NCBI requires an email for API access.

EMAIL="your.email@example.com"

# ===== 12S rRNA (MiFish-U marker) =====
crabs --download-ncbi --database nucleotide --query 'Actinopterygii[Organism] AND ("12S ribosomal RNA"[Title] OR 12S[Gene])' --email $EMAIL --output raw_data/ncbi_12s_actinopterygii.fasta
crabs --download-ncbi --database nucleotide --query 'Chondrichthyes[Organism] AND ("12S ribosomal RNA"[Title] OR 12S[Gene])' --email $EMAIL --output raw_data/ncbi_12s_chondrichthyes.fasta
crabs --download-ncbi --database nucleotide --query 'Sarcopterygii[Organism] AND ("12S ribosomal RNA"[Title] OR 12S[Gene])' --email $EMAIL --output raw_data/ncbi_12s_sarcopterygii.fasta
crabs --download-ncbi --database nucleotide --query 'Myxini[Organism] AND ("12S ribosomal RNA"[Title] OR 12S[Gene])' --email $EMAIL --output raw_data/ncbi_12s_myxini.fasta
crabs --download-ncbi --database nucleotide --query 'Petromyzontidae[Organism] AND ("12S ribosomal RNA"[Title] OR 12S[Gene])' --email $EMAIL --output raw_data/ncbi_12s_lampreys.fasta
crabs --download-ncbi --database nucleotide --query 'Cetacea[Organism] AND ("12S ribosomal RNA"[Title] OR 12S[Gene])' --email $EMAIL --output raw_data/ncbi_12s_cetacea.fasta
crabs --download-ncbi --database nucleotide --query 'Pinnipedia[Organism] AND ("12S ribosomal RNA"[Title] OR 12S[Gene])' --email $EMAIL --output raw_data/ncbi_12s_pinnipedia.fasta

# ===== 16S rRNA (STAT / MarVer markers) =====
crabs --download-ncbi --database nucleotide --query '(Cetacea[Organism] AND 16S[All Fields]) AND 1000:2000[Sequence Length]' --email $EMAIL --output raw_data/ncbi_16s_cetacea.fasta
crabs --download-ncbi --database nucleotide --query '(Sirenia[Organism] AND 16S[All Fields]) AND 1000:2000[Sequence Length]' --email $EMAIL --output raw_data/ncbi_16s_sirenia.fasta
crabs --download-ncbi --database nucleotide --query '("Pinnipedia"[Organism] AND 16S[All Fields]) AND 1000:2000[Sequence Length]' --email $EMAIL --output raw_data/ncbi_16s_pinnipedia.fasta

# ===== COI (Leray-XT marker) =====
crabs --download-ncbi --database nucleotide --query '("Actinopterygii"[Organism] AND ("COX1"[Gene] OR "cytochrome c oxidase subunit I"[Gene])' --email $EMAIL --output raw_data/ncbi_coi_actinopterygii.fasta
crabs --download-ncbi --database nucleotide --query '("Chondrichthyes"[Organism] AND ("COX1"[Gene] OR "cytochrome c oxidase subunit I"[Gene]))' --email $EMAIL --output raw_data/ncbi_coi_chondrichthyes.fasta
crabs --download-ncbi --database nucleotide --query '("Sarcopterygii"[Organism] AND ("COX1"[Gene] OR "cytochrome c oxidase subunit I"[Gene]))' --email $EMAIL --output raw_data/ncbi_coi_sarcopterygii.fasta
crabs --download-ncbi --database nucleotide --query '("Cetacea"[Organism] AND ("COX1"[Gene] OR "cytochrome c oxidase subunit I"[Gene]))' --email $EMAIL --output raw_data/ncbi_coi_cetacea.fasta
crabs --download-ncbi --database nucleotide --query '("Pinnipedia"[Organism] AND ("COX1"[Gene] OR "cytochrome c oxidase subunit I"[Gene]))' --email $EMAIL --output raw_data/ncbi_coi_pinnipedia.fasta

2.3 Download from EMBL

# Vertebrate mitochondrial sequences
crabs --download-embl --taxon 'VRT_\d+\' --output raw_data/embl_vertebrates.fasta

# Mammal mitochondrial sequences
crabs --download-embl --taxon 'MAM_\d+\' --output raw_data/embl_mammals.fasta

2.4 Download from MitoFish

Recommended: manual download

1. Navigate to mitofish.aori.u-tokyo.ac.jp/download.html
2. Download Complete + Partial Mitochondrial Genomes
3. Save as raw_data/mitofish_complete.fasta

Or use automated download:

crabs --download-mitofish --output raw_data/mitofish_complete.fasta

2.5 Verify Downloads

ls -lh raw_data/

Expected: ~27–30 FASTA files (BOLD: ~7, NCBI: ~17, EMBL: ~2, MitoFish: ~1)

---

Phase 3: Import & Merge Sequences

All sequences must be converted to CRABS format with standardised NCBI taxonomy before processing.

3.1 Import BOLD Sequences

for taxon in actinopterygii_coi chondrichthyes_coi sarcopterygii_coi myxini_coi lampreys_coi cetacea_coi; do
  crabs --import \
    --import-format boldv3 \
    --input raw_data/bold_${taxon}.fasta \
    --names taxonomy/names.dmp \
    --nodes taxonomy/nodes.dmp \
    --acc2tax taxonomy/nucl_gb.accession2taxid \
    --output processed_data/bold_${taxon}.txt \
    --ranks 'superkingdom;phylum;class;order;family;genus;species'
done

Note

Skip import commands for any files that don't exist or contain no sequences (e.g., Myxini, Sarcopterygii may have limited BOLD data).

3.2 Import NCBI Sequences

for file in ncbi_12s_actinopterygii ncbi_12s_chondrichthyes ncbi_12s_sarcopterygii ncbi_12s_myxini ncbi_12s_lampreys ncbi_12s_cetacea ncbi_12s_pinnipedia ncbi_16s_cetacea ncbi_16s_pinnipedia ncbi_16s_sirenia ncbi_coi_actinopterygii ncbi_coi_chondrichthyes ncbi_coi_sarcopterygii ncbi_coi_cetacea ncbi_coi_pinnipedia; do
  crabs --import \
    --import-format ncbi \
    --input raw_data/${file}.fasta \
    --names taxonomy/names.dmp \
    --nodes taxonomy/nodes.dmp \
    --acc2tax taxonomy/nucl_gb.accession2taxid \
    --output processed_data/${file}.txt \
    --ranks 'superkingdom;phylum;class;order;family;genus;species'
done

3.3 Import EMBL Sequences

for source in vertebrates mammals; do
  crabs --import \
    --import-format embl \
    --input raw_data/embl_${source}.fasta \
    --names taxonomy/names.dmp \
    --nodes taxonomy/nodes.dmp \
    --acc2tax taxonomy/nucl_gb.accession2taxid \
    --output processed_data/embl_${source}.txt \
    --ranks 'superkingdom;phylum;class;order;family;genus;species'
done

3.4 Import MitoFish Sequences

crabs --import \
  --import-format mitofish \
  --input raw_data/mitofish_complete.fasta \
  --names taxonomy/names.dmp \
  --nodes taxonomy/nodes.dmp \
  --acc2tax taxonomy/nucl_gb.accession2taxid \
  --output processed_data/mitofish_complete.txt \
  --ranks 'superkingdom;phylum;class;order;family;genus;species'

Checkpoint: You should have 10+ .txt files in processed_data/.

---

Phase 4–6: Marker-Specific Workflows

Each marker follows the same six steps: Merge → In Silico PCR → Filter → Dereplicate → Subset → Export

---

Workflow A: MiFish-U (12S rRNA)

Target: Universal fish and cetacean detection | Amplicon: 160–180 bp

A1 — Merge 12S Sources

crabs --merge \
  --output processed_data/merged_12s.txt \
  --import-format boldv5 \
  --uniq \
  --input "processed_data/ncbi_12s_actinopterygii.txt;
           processed_data/ncbi_12s_chondrichthyes.txt;
           processed_data/ncbi_12s_sarcopterygii.txt;
           processed_data/ncbi_12s_myxini.txt;
           processed_data/ncbi_12s_lampreys.txt;
           processed_data/ncbi_12s_cetacea.txt;
           processed_data/ncbi_12s_pinnipedia.txt;
           processed_data/mitofish_complete.txt;
           processed_data/embl_vertebrates.txt;
           processed_data/embl_mammals.txt"

Expected: ~500,000–2,000,000 sequences (~2–4 GB)

A2 — In Silico PCR

crabs --in-silico-pcr \
  --input processed_data/merged_12s.txt \
  --output processed_data/mifish_amplicons.txt \
  --forward GTCGGTAAAACTCGTGCCAGC \
  --reverse CATAGTGGGGTATCTAATCCCAGTTTG \
  --mismatch 4 \
  --buffer-size 988978900

OverflowError fix

If you get a buffer overflow error, calculate the correct buffer size with:

awk -F'\t' '{ print length($NF) }' processed_data/merged_12s.txt | sort -nr | head -n 1 | awk '{ print $1*2 }'

Expected: ~50,000–200,000 sequences

A3 — Quality Filter

crabs --filter \
  --input processed_data/mifish_amplicons.txt \
  --output processed_data/mifish_filtered.txt \
  --minimum-length 100 \
  --maximum-length 250 \
  --maximum-n 2 \
  --environmental

A4 — Dereplicate

crabs --dereplicate \
  --input processed_data/mifish_filtered.txt \
  --output processed_data/mifish_derep.txt \
  --dereplication-method unique_species

Expected: ~8,000–30,000 sequences (50–80% reduction)

A5 — Subset to South African Species

crabs --subset \
  --input processed_data/mifish_derep.txt \
  --output processed_data/mifish_sa_subset.txt \
  --include merged_sa_species.txt

# Check how many SA species have 12S sequences
cut -f7 processed_data/mifish_sa_subset.txt | sort -u | wc -l

Expected: ~1,000–5,000 sequences

A6 — Export

crabs --export --input processed_data/mifish_sa_subset.txt --output reference_dbs/mifish_qiime.fasta --export-format qiime-fasta
crabs --export --input processed_data/mifish_sa_subset.txt --output reference_dbs/mifish_blast.fasta --export-format blast
crabs --export --input processed_data/mifish_sa_subset.txt --output reference_dbs/mifish_sintax.fasta --export-format sintax

---

Workflow B: MarVer (Marine Mammals — 12S & 16S)

Target: Cetaceans, Pinnipeds, Sirenians | Markers: MarVer1 (12S) + MarVer3 (16S)

B1 — Merge Marine Mammal Sources

crabs --merge \
  --output processed_data/marine_mammals_16s_merged.txt \
  --uniq \
  --input "processed_data/ncbi_16s_cetacea.txt;
           processed_data/ncbi_16s_pinnipedia.txt;
           processed_data/ncbi_16s_sirenia.txt"

Why not include EMBL MAM?

EMBL MAM contains all mammals — 96% are terrestrial primates. STAT/MarVer primers are marine mammal-specific and won't amplify terrestrial species, so including them wastes processing time.

B2 — In Silico PCR (both primers)

crabs --in-silico-pcr \
  --input processed_data/marine_mammals_16s_merged.txt \
  --output processed_data/amplicons_MarVer1.txt \
  --forward CGTGCCAGCCACCGCG \
  --reverse GGGTATCTAATCCYAGTTTG \
  --mismatch 4.5

crabs --in-silico-pcr \
  --input processed_data/marine_mammals_16s_merged.txt \
  --output processed_data/amplicons_MarVer3.txt \
  --forward AGACGAGAAGACCCTRTG \
  --reverse GGATTGCGCTGTTATCCC \
  --mismatch 4.5

B3 — Tag and Merge Amplicons

# Add primer tags to accession IDs to distinguish sources after merging
awk -F'\t' 'BEGIN {OFS="\t"} {$1=$1"_MarVer1"; print}' processed_data/amplicons_MarVer1.txt > processed_data/amplicons_MarVer1_tagged.txt
awk -F'\t' 'BEGIN {OFS="\t"} {$1=$1"_MarVer3"; print}' processed_data/amplicons_MarVer3.txt > processed_data/amplicons_MarVer3_tagged.txt

cat processed_data/amplicons_MarVer1_tagged.txt processed_data/amplicons_MarVer3_tagged.txt > processed_data/amplicons_combined_tagged.txt

B4 — Quality Filter

crabs --filter \
  --input processed_data/amplicons_combined_tagged.txt \
  --output processed_data/amplicons_cleaned.txt \
  --minimum-length 100 \
  --maximum-length 500 \
  --maximum-n 0

B5 — Dereplicate

crabs --dereplicate \
  --input processed_data/amplicons_cleaned.txt \
  --output processed_data/amplicons_dereplicated.txt \
  --dereplication-method unique_species

B6 — Subset to SA Species

crabs --subset \
  --input processed_data/amplicons_dereplicated.txt \
  --output processed_data/amplicons_sa_subset.txt \
  --include merged_sa_species.txt

B7 — Export

crabs --export --input processed_data/amplicons_sa_subset.txt --output reference_dbs/marine_mammals_qiime.fasta --export-format qiime-fasta
crabs --export --input processed_data/amplicons_sa_subset.txt --output reference_dbs/marine_mammals_blast.fasta --export-format blast
crabs --export --input processed_data/amplicons_sa_subset.txt --output reference_dbs/marine_mammals_sintax.fasta --export-format sintax

---

Workflow C: Leray-XT (COI)

Target: Fish, cetaceans, invertebrates | Amplicon: ~313 bp | Largest dataset

C1 — Merge COI Sources

crabs --merge \
  --output processed_data/merged_coi.txt \
  --uniq \
  --input "processed_data/bold_actinopterygii_coi.txt;
           processed_data/bold_chondrichthyes_coi.txt;
           processed_data/bold_sarcopterygii_coi.txt;
           processed_data/bold_myxini_coi.txt;
           processed_data/bold_lampreys_coi.txt;
           processed_data/bold_cetacea_coi.txt;
           processed_data/ncbi_coi_actinopterygii.txt;
           processed_data/ncbi_coi_chondrichthyes.txt;
           processed_data/ncbi_coi_sarcopterygii.txt;
           processed_data/ncbi_coi_cetacea.txt;
           processed_data/ncbi_coi_pinnipedia.txt;
           processed_data/mitofish_complete.txt;
           processed_data/embl_vertebrates.txt"

Expected: ~1,000,000–3,000,000 sequences (~5–10 GB)

C2 — In Silico PCR

# Calculate buffer size first
awk -F'\t' '{ print length($NF) }' processed_data/merged_coi.txt | sort -nr | head -n 1 | awk '{ print $1*2 }'

crabs --in-silico-pcr \
  --input processed_data/merged_coi.txt \
  --output processed_data/leray_amplicons.txt \
  --forward GGWACWRGWTGRACWITITAYCCYCC \
  --reverse TAIACYTCIGGRTGICCRAARAAYCA \
  --mismatch 4 \
  --buffer-size 452178200

C3 — Quality Filter

crabs --filter \
  --input processed_data/leray_amplicons.txt \
  --output processed_data/leray_filtered.txt \
  --minimum-length 250 \
  --maximum-length 400 \
  --maximum-n 2 \
  --environmental

C4 — Dereplicate

crabs --dereplicate \
  --input processed_data/leray_filtered.txt \
  --output processed_data/leray_derep.txt \
  --dereplication-method unique_species

C5 — Subset to SA Species

crabs --subset \
  --input processed_data/leray_derep.txt \
  --output processed_data/leray_sa_subset.txt \
  --include merged_sa_species.txt

Expected: ~3,000–15,000 sequences

C6 — Export

crabs --export --input processed_data/leray_sa_subset.txt --output reference_dbs/leray_qiime.fasta --export-format qiime-fasta
crabs --export --input processed_data/leray_sa_subset.txt --output reference_dbs/leray_blast.fasta --export-format blast
crabs --export --input processed_data/leray_sa_subset.txt --output reference_dbs/leray_sintax.fasta --export-format sintax

---

Phase 7: Create BLAST Databases

makeblastdb -in reference_dbs/mifish_blast.fasta -dbtype nucl -parse_seqids -out reference_dbs/mifish_blastdb
makeblastdb -in reference_dbs/marine_mammals_blast.fasta -dbtype nucl -parse_seqids -out reference_dbs/marine_mammals_blastdb
makeblastdb -in reference_dbs/leray_blast.fasta -dbtype nucl -parse_seqids -out reference_dbs/leray_blastdb

Each marker should produce .nhr, .nin, and .nsq index files.

---

Phase 8: Visualisation & Quality Control

Database Diversity Figures

# --tax-level: 1=superkingdom, 2=phylum, 3=class, 4=order, 5=family, 6=genus, 7=species
crabs --diversity-figure --input processed_data/mifish_sa_subset.txt --output results/mifish_diversity.png --tax-level 4
crabs --diversity-figure --input processed_data/amplicons_sa_subset.txt --output results/marver_diversity.png --tax-level 4
crabs --diversity-figure --input processed_data/leray_sa_subset.txt --output results/leray_diversity.png --tax-level 4

Amplicon Length Distributions

crabs --amplicon-length-figure --input processed_data/mifish_sa_subset.txt --tax-level 4 --output results/mifish_lengths.png
crabs --amplicon-length-figure --input processed_data/leray_sa_subset.txt --tax-level 4 --output results/leray_lengths.png

Species Coverage Report

echo "MiFish-U (12S) species:"; cut -f7 processed_data/mifish_sa_subset.txt | sort -u | wc -l
echo "MarVer (16S/12S) species:"; cut -f7 processed_data/amplicons_sa_subset.txt | sort -u | wc -l
echo "Leray-XT (COI) species:"; cut -f7 processed_data/leray_sa_subset.txt | sort -u | wc -l
echo "Total SA species in list:"; wc -l merged_sa_species.txt

# Find species covered by both 12S and COI
cut -f7 processed_data/mifish_sa_subset.txt | sort -u > results/mifish_species_list.txt
cut -f7 processed_data/leray_sa_subset.txt | sort -u > results/leray_species_list.txt
comm -12 results/mifish_species_list.txt results/leray_species_list.txt > results/species_in_both_12S_COI.txt
wc -l results/species_in_both_12S_COI.txt

---

Phase 9: Downstream Analysis

BLAST Assignment

blastn \
  -query your_asvs.fasta \
  -db reference_dbs/mifish_blastdb \
  -out results/blast_results.txt \
  -outfmt '6 qseqid sseqid pident length qcovs evalue bitscore staxids sscinames scomnames' \
  -max_target_seqs 5 \
  -perc_identity 97 \
  -num_threads 4

DADA2 Assignment (R)

library(dada2)

taxa <- assignTaxonomy(
  seqs = seqtab.nochim,
  refFasta = "reference_dbs/mifish_qiime.fasta",
  minBoot = 80,
  multithread = TRUE,
  tryRC = TRUE
)
write.csv(taxa, "results/mifish_dada2_taxonomy.csv")

QIIME2 Assignment

conda activate qiime2-amplicon-2024.10

# Import reference sequences
qiime tools import \
  --type 'FeatureData[Sequence]' \
  --input-path reference_dbs/mifish_qiime.fasta \
  --output-path reference_dbs/mifish_seqs.qza

# Train classifier
qiime feature-classifier fit-classifier-naive-bayes \
  --i-reference-reads reference_dbs/mifish_seqs.qza \
  --i-reference-taxonomy reference_dbs/mifish_taxonomy.qza \
  --o-classifier reference_dbs/mifish_classifier.qza

# Classify
qiime feature-classifier classify-sklearn \
  --i-classifier reference_dbs/mifish_classifier.qza \
  --i-reads your_rep_seqs.qza \
  --o-classification results/mifish_taxonomy.qza

---

Troubleshooting

Problem	Solution
FileNotFoundError: taxonomy/names.dmp	Re-download taxonomy files (Phase 1.3)
In silico PCR returns < 1,000 sequences	Increase --mismatch to 5 or 6; verify primer sequences
OverflowError: FASTA record does not fit into buffer	Recalculate --buffer-size (see Phase A2 tip above)
Subset returns empty file	Check species name format matches between database and merged_sa_species.txt (e.g., Genus species not Genus_species)
MemoryError during merge	Process taxa groups separately, then merge the subsetted results
BLAST database creation fails	Run makeblastdb -version; install with conda install -c bioconda blast

---

Summary Checklist

• [ ] Phase 1: Conda environment active; taxonomy files downloaded
• [ ] Phase 2: All sequences downloaded (BOLD, NCBI, EMBL, MitoFish)
• [ ] Phase 3: All sequences imported to CRABS format
• [ ] Workflow A (MiFish-U): Merged → PCR → Filtered → Dereplicated → Subsetted → Exported → BLAST DB
• [ ] Workflow B (MarVer): Merged → PCR → Tagged → Filtered → Dereplicated → Subsetted → Exported → BLAST DB
• [ ] Workflow C (Leray-XT): Merged → PCR → Filtered → Dereplicated → Subsetted → Exported → BLAST DB
• [ ] Phase 8: Diversity figures and species coverage reports generated
• [ ] Phase 9: Ready for taxonomic assignment (BLAST / DADA2 / QIIME2)

---

Final Directory Structure

edna_sa_project/
├── merged_sa_species.txt        (2,087 species)
├── taxonomy/
│   ├── names.dmp                (272 MB)
│   ├── nodes.dmp                (205 MB)
│   └── nucl_gb.accession2taxid  (13 GB)
├── raw_data/                    (~27–30 FASTA files)
├── processed_data/              (~20+ CRABS .txt files)
├── reference_dbs/
│   ├── mifish_qiime.fasta       mifish_blast.fasta   mifish_sintax.fasta
│   ├── mifish_blastdb.nhr/nin/nsq
│   ├── marine_mammals_*.fasta   + BLAST index files
│   └── leray_*.fasta            + BLAST index files
└── results/
    ├── *_diversity.png
    ├── *_lengths.png
    └── *_species_list.txt

---

References

Jeunen G-J, Dowle E, Edgecombe J, von Ammon U, Gemmell NJ, Cross H (2022). crabs — A software program to generate curated reference databases for metabarcoding sequencing data. Molecular Ecology Resources. doi:10.1111/1755-0998.13741

Valsecchi E et al. (2020). Novel universal primers for metabarcoding environmental DNA surveys of marine mammals and other marine vertebrates. Environmental DNA, 2(4), 602–617.

• CRABS GitHub
• FishBase
• SeaLifeBase