# Error handling.
set -ue

# This function summarizes statistics in a clustring filr
function stats() {
    # Prints minimum, maximum and median cluster similarity across all clusters
    grep at | awk ' { print $5 } ' | tr -d '%' | datamash min 1 max 1 median 1
}

# Cluster the sequences.
cd-hit -i refs./sars1.fa -o out >> log

# Print the average similarities.
cat out.clstr | stats