Glossary

Pharmacogenomics (PGx) is the study of how an individual's genetic makeup influences their response to medications. It combines pharmacology (the science of drugs) and genomics (the study of genes and their functions) to develop effective, personalized drug therapies based on a patient's DNA.

A single nucleotide polymorphism (SNP, pronounced 'snip') is a variation at a single position in a DNA sequence among individuals. SNPs are the most common type of genetic variation in humans, with approximately 4-5 million SNPs per individual genome and over 660 million cataloged in the dbSNP database.

An rsID (reference SNP cluster ID) is a unique identifier assigned by the NCBI dbSNP database to a specific genetic variant, typically a single nucleotide polymorphism (SNP). The format is 'rs' followed by a number — for example, rs429358 identifies the APOE4-defining variant. rsIDs serve as the universal language for referencing genetic variants across research, clinical testing, and bioinformatics tools.

Fmoc-SPPS (Fluorenylmethyloxycarbonyl Solid-Phase Peptide Synthesis) is the dominant chemical method for synthesizing peptides in both research and pharmaceutical manufacturing. The peptide chain is assembled from C-terminus to N-terminus on an insoluble resin support, with each amino acid's alpha-amino group protected by the Fmoc group, which is removed with piperidine before coupling the next residue.

ESMFold is a protein structure prediction model developed by Meta AI (formerly Facebook AI Research) that predicts the three-dimensional structure of a protein directly from its amino acid sequence. Unlike AlphaFold, ESMFold does not require multiple sequence alignments (MSAs), enabling predictions in seconds rather than minutes, which makes it particularly suitable for high-throughput peptide design pipelines.

pLDDT (predicted Local Distance Difference Test) is a per-residue confidence metric produced by protein structure prediction models such as AlphaFold2 and ESMFold. It estimates how accurately each amino acid's position has been predicted, scored on a scale from 0 to 100, where higher values indicate greater confidence in the predicted local structure.

ClinVar is a freely accessible public database maintained by the National Center for Biotechnology Information (NCBI) that aggregates information about the relationships between human genetic variants and observed health conditions (phenotypes). Submitters — including clinical laboratories, research groups, and expert panels — classify variants using a standardized five-tier system: pathogenic, likely pathogenic, uncertain significance (VUS), likely benign, and benign.

UniProt (Universal Protein Resource) is the most comprehensive, freely accessible database of protein sequence and functional information. It is maintained by a consortium of the European Bioinformatics Institute (EMBL-EBI), the Swiss Institute of Bioinformatics (SIB), and the Protein Information Resource (PIR). UniProt contains over 250 million protein sequences, with its curated section (Swiss-Prot) providing expert-reviewed annotations for approximately 570,000 proteins.

Peptide therapeutics are a class of pharmaceutical drugs composed of short chains of amino acids, typically between 2 and 50 residues in length. They occupy a unique niche between small-molecule drugs and large biologic proteins, combining the target specificity of antibodies with improved tissue penetration and lower manufacturing costs. The global peptide therapeutics market exceeded $50 billion in 2023 and is projected to grow at 9-10% annually.

GLP-1 receptor agonists (GLP-1 RAs) are a class of peptide drugs that mimic the incretin hormone glucagon-like peptide-1. They bind to and activate the GLP-1 receptor on pancreatic beta cells, stimulating glucose-dependent insulin secretion, suppressing glucagon release, slowing gastric emptying, and reducing appetite. Major examples include semaglutide (Ozempic, Wegovy, Rybelsus), liraglutide (Victoza, Saxenda), and tirzepatide (Mounjaro, Zepbound).

Personalized medicine (also called precision medicine) is a medical model that uses an individual's genetic, environmental, and lifestyle information to guide clinical decisions. Rather than prescribing the same drug at the same dose to every patient with a given condition, personalized medicine selects therapies and dosages based on the patient's unique biological profile — particularly their genomic data.

Cytochrome P450 (CYP450) enzymes are a superfamily of heme-containing monooxygenases that catalyze the oxidative metabolism of the majority of clinically used drugs. In humans, 57 CYP genes encode enzymes that metabolize endogenous substrates (steroids, bile acids, fatty acids) and xenobiotics (drugs, environmental chemicals, dietary compounds). Five CYP enzymes — CYP1A2, CYP2C9, CYP2C19, CYP2D6, and CYP3A4 — are responsible for metabolizing approximately 90% of all drugs in clinical use.

A poor metabolizer (PM) is an individual who carries genetic variants resulting in little or no functional activity of a drug-metabolizing enzyme, most commonly a cytochrome P450 (CYP450) enzyme. Poor metabolizers process certain drugs much more slowly than normal metabolizers, which can lead to drug accumulation, increased plasma levels, prolonged drug effects, and a higher risk of adverse drug reactions at standard doses.

Binding affinity is a quantitative measure of the strength of interaction between two molecules, typically a drug (ligand) and its biological target (receptor or protein). It is most commonly expressed as the dissociation constant (Kd), which represents the concentration of ligand at which 50% of the target binding sites are occupied. A lower Kd indicates stronger binding — nanomolar (nM) or picomolar (pM) affinities are typical for effective drugs.

De novo peptide design is the computational creation of novel peptide sequences that do not exist in nature, engineered from scratch to achieve specific therapeutic objectives. Unlike peptide discovery from natural sources (venoms, hormones, antimicrobial peptides), de novo design uses algorithms, molecular modeling, and machine learning to generate sequences optimized for target binding, stability, selectivity, and manufacturability.