Why do we each have unique microbiomes? What are the consequences of individuality for therapeutic design?

The human microbiome is remarkably personalized – even people living together harbor distinct microbial communities. On the skin, individuals in a family often share the same species yet harbor distinct but dynamic strain-level communities. This personalization may explain why most microbiome therapies fail to consistently engraft across patients.

The Lieberman Lab seeks to understand how ecology and evolution shape these personalized communities, and the role of this personalization on human health.

Area I: The interplay of rapid evolution and ecology in complex microbial environments.

Bacterial adaptations with complex communities can be rapid (Zhao et al 2019, Key et al 2023, Poret et al 2025). Tracking adaptations can inform therapeutic design, and understanding the tempo and nature of adaption can reveal foundational principles for microbiome community assembly.

In-person adaptations have signatures of parallelism, person-specificity, and tradeoffs (Lieberman 2022). Our recent theoretical work has suggested that these adaptations are commonly missed by classic evolutionary analyses (Torrillo & Lieberman 2024).

Similarly, understanding ecological interactions is critical to the design of microbiome-based therapies. For example, our recent demonstration that intra-species warfare is ecologically relevant in determining colonization in the skin microbiome (Mancuso et al 2025) suggests that microbiome therapies must account for antimicrobials produced by the strains currently on a person.

Click to reveal ongoing and upcoming projects in this area.

• 🧪🧪💻 💻 Understanding how resistance to warfare molecules evolves and the tradeoffs that maintain sensitivity. This work extends Mancuso et al 2025. 🧪
  
• 💻💻💻💻 Investigating evidence of past reversions in genomes and gut microbiomes. This work extends Torrillo & Lieberman 2024 and Poret et al 2025.
  
• 🧪💻💻💻 Comparing patterns of transmission and evolution across species with diverse lifestyles in the skin microbiome, to understand the rules that determine eco-evo dynamics. This work extends Baker et al 2025.
  
• 🧪🧪💻 💻 Tracking evolution following probiotic administration in the gut microbiome and vaginal microbiome.

Area II : Host-microbe interactions in the skin microbiome

We seek out opportunities where taking a data-science or evolutionary approach can advance understanding of the roles of microbes in skin diseases. We have contributed to understanding the role of microbes during wound healing in mice (Khadka and Markey et al 2024) and in humans (Gupta and Poret et al 2024).

We have also studied the microbiome in atopic dermatitis (Khadka and Key et al 2021), including the finding that S. aureus continually adapts to the skin of individuals with atopic dermatitis (Key et al 2023). This discovery raises the possibility that sustained colonization and continual adaptation during complex disease may exacerbate the positive-feedback cycle between pathogen growth and barrier damage.

Going forward, we are particularly interested in the role C. acnes plays in health and disease. Click to reveal ongoing and upcoming projects

• 🧪🧪💻💻 Tracking trait evolution across the Cutibacterium phylogeny to identify traits associated with strain success on different hosts. This work follows up on results from Qu et al 2025.
  
• 🧪🧪🧪💻 Developing new methods for single-cell and spatial transcriptomics and applying them to study Cutibacterium acnes transcriptomics in vivo.
  
• 🧪 or 💻 Exploring the role of resident microbiota in preventing skin infection. This work follows on Gupta and Poret et al 2024.
  
• 🧪🧪🧪💻 Revisiting the role of C. acnes in acne vulgaris using novel sampling and sequencing approaches.

Area III: Models and machine learning for evolution and microbiome analyses

We develop computational tools and theoretical models to both improve data analysis and understand microbial community assembly in the real world. We have developed computational tools for precision strain calling from difficult sample types (PHLAME, Qu et al 2025) and prevision mutation calling from large isolate collections (AccuSNV, Liao et al, in prep), as well as theoretical models to understand possible evolutionary scenarios from available metagenomic data (Torrillo & Lieberman 2024).

We are currently embracing the very large publicly available data sets available in microbial genomics (>100,000 high quality genomes of a single species) and metagenomics (>100,000 genomes across environments) and modern machine learning techniques to ask applied and fundamental questions in microbiology.

Click to reveal ongoing and upcoming projects

• 💻💻💻💻 Machine learning to predict which strains are most likely to acquire a gene of interest, and using the underlying model to understand gene-gene interactions.
  
• 💻💻💻💻 Machine learning to predict colonization within gut microbiomes.

Advantages of studying human skin microbiomes:

We study a variety of human-associated and environmental microbiomes, including the gut, skin and vaginal microbiomes. While our largest focus is currently on the skin microbiome, we start with interesting questions across human microbiomes and choose the best system for the question. The skin is an ideal system for many basic questions for several reasons:

In addition, many human skin diseases to not have good animal models; we hope to uncover the mechanisms of skin diseases, including atopic dermatitis and acne while asking fundamental questions about microbiome communities.

Strategies

We strive to:

• Address questions without clear answers in all real-world microbial communities
• Enable strong inferences and clear answers by gathering the highest quality data
• Choose approaches that are as complex as required, but no more complex