Analyzing inherent biases in SARS-CoV-2 PCR and serological epidemiologic metrics