NCBI Assembly Reports

Preparations

cbp install nwr
cbp install sqlite3
cbp install tva

Requires SQLite version 3.34 or above. sqlite that comes with mac does not work.

NCBI Taxonomy Statistics

curl -L "https://www.ncbi.nlm.nih.gov/Taxonomy/taxonomyhome.html/index.cgi?chapter=statistics&?&unclassified=hide&uncultured=hide" |
    tva from html -q 'table[bgcolor="#CCCCFF"] table[bgcolor="#FFFFFF"] tr td text{}' |
    grep '\S' |
    paste -d $'\t' - - - - - - |
    tva to md --right 2-6

NCBI ASSEMBLY

• assembly_level

for C in refseq genbank; do
    cat ~/.nwr/assembly_summary_${C}.txt |
        sed '1d' |
        tva stats -H -g assembly_level,genome_rep --count |
        tva keep-header -- sort |
        tva to md --fmt

    echo -e "\nTable: ${C}\n\n"
done

Table: refseq

Table: genbank

Example 1: count qualified assemblies of Eukaryote groups

ARRAY=(
    # Animals - Metazoa - kingdom
    'Flatworms::Platyhelminthes' # phylum
    'Roundworms::Nematoda'
    'Insects::Hexapoda' # subphylum
    'Reptiles::Testudines' # order
    'Reptiles::Lepidosauria'
    'Reptiles::Crocodylia'
    'Fishes::Chondrichthyes' # class
    'Fishes::Dipnoi'
    'Fishes::Actinopterygii'
    'Fishes::Hyperotreti'
    'Fishes::Hyperoartia'
    'Fishes::Coelacanthimorpha'
    'Mammals::Mammalia'
    'Birds::Aves'
    'Amphibians::Amphibia'
    # Fungi - kindom
    'Ascomycetes::Ascomycota' # phylum
    'Basidiomycetes::Basidiomycota'
    # Plants - Viridiplantae
    'Green Plants::Viridiplantae'
    'Land Plants::Embryophyta'
    # Protists
    'Apicomplexans::Apicomplexa'
    'Kinetoplasts::Kinetoplastida'
)

echo -e "GROUP_NAME\tSCI_NAME\tComplete Genome\tChromosome\tScaffold\tContig" \
    > groups.tsv

for item in "${ARRAY[@]}" ; do
    GROUP_NAME="${item%%::*}"
    SCI_NAME="${item##*::}"

    GENUS=$(
        nwr member ${SCI_NAME} -r genus |
            grep -v -i "Candidatus " |
            grep -v -i "candidate " |
            sed '1d' |
            cut -f 1 |
            tr "\n" "," |
            sed 's/,$/\)/' |
            sed 's/^/\(/'
    )

    printf "$GROUP_NAME\t$SCI_NAME\t"

    for L in 'Complete Genome' 'Chromosome' 'Scaffold' 'Contig'; do
        echo "
            SELECT
                COUNT(*)
            FROM ar
            WHERE 1=1
                AND genus_id IN $GENUS
                AND assembly_level IN ('$L')
            " |
            sqlite3 -tabs ~/.nwr/ar_refseq.sqlite
    done |
    tr "\n" "\t" |
    sed 's/\t$//'

    echo;
done \
    >> groups.tsv

cat groups.tsv |
    tva to md --num

Table: refseq - Eukaryotes

Table: genbank - Eukaryotes

Example 2: find accessions of a species

Staphylococcus capitis - 29388 - 头状葡萄球菌

nwr info "Staphylococcus capitis"

nwr member 29388

echo '
.headers ON
    SELECT
        organism_name,
        species,
        genus,
        ftp_path,
        assembly_level
    FROM ar
    WHERE 1=1
        AND tax_id != species_id    -- with strain ID
        AND species_id IN (29388)
    ' |
    sqlite3 -tabs ~/.nwr/ar_refseq.sqlite \
    > Scap.assembly.tsv

echo '
    SELECT
        species || " " || REPLACE(assembly_accession, ".", "_") AS organism_name,
        species,
        genus,
        ftp_path,
        assembly_level
    FROM ar
    WHERE 1=1
        AND tax_id = species_id     -- no strain ID
        AND assembly_level IN ("Chromosome", "Complete Genome")
        AND species_id IN (29388)
    ' |
    sqlite3 -tabs ~/.nwr/ar_refseq.sqlite \
    >> Scap.assembly.tsv

Example 3: find model organisms in a family

echo "
.headers ON
    SELECT
        tax_id,
        organism_name
    FROM ar
    WHERE 1=1
        AND family IN ('Enterobacteriaceae')
        AND refseq_category IN ('reference genome')
    " |
    sqlite3 -tabs ~/.nwr/ar_refseq.sqlite |
    sed '1s/^/#/' |
    tva to md