Lineage tracing is a set of techniques used to follow a single cell and all of its descendants over time, building a family tree of how one cell becomes many. The basic idea sounds deceptively simple: tag a cell, wait, and see what it turns into. But making that work inside a living organism, where billions of cells are dividing, migrating, and dying in three-dimensional tissue, has required more than a century of increasingly clever biology. The result is a toolkit that now underpins major discoveries in stem cell science, cancer research, organ regeneration, and embryonic development.
From Watching Cells to Marking Them
The earliest forms of lineage tracing were purely observational. Researchers in the late 1800s and early 1900s watched transparent embryos under the microscope, physically tracking individual cells as they divided. This worked well in small, see-through organisms but broke down the moment tissues became opaque or complex. The field evolved by developing ways to mark cells so their progeny could be identified later, even deep inside an animal. Transplant experiments, in which tissue from one species was grafted into another, allowed researchers to distinguish donor-derived cells by appearance. These classical approaches laid the foundation, but they were laborious, limited in scale, and often impossible in mammals.
The leap to genetic marking changed everything. In the 1980s and 1990s, researchers began inserting reporter genes, such as one encoding a detectable enzyme, into cells using engineered viruses. By infecting a tissue at low levels so that each virus landed in a different cell, they could trace the descendants of individual progenitors. This retroviral approach was used to map how progenitor cells in the developing rodent brain gave rise to diverse neuron types, establishing that single progenitors could be multipotent.1Cell Press. Evolving strategies for lineage tracing – Section: Introduction It was a conceptual breakthrough: for the first time, you could mark a cell and then read its history from its DNA rather than from direct observation.
The Cre-Lox System and Inducible Marking
The workhorse of modern lineage tracing in mice is a genetic system built around a bacterial enzyme called Cre recombinase and short DNA sequences called lox sites. The setup works like a permanent molecular switch. When Cre is active in a cell, it snips out a segment of DNA flanked by lox sites, which can activate a reporter gene, say one that makes a fluorescent protein. Once the DNA is cut, the change is irreversible and inherited by every daughter cell. Even if the original signal that triggered Cre disappears, the reporter stays on permanently, marking the entire lineage.
What makes this system especially powerful is that Cre can be placed under the control of a gene that is active only in a specific cell type. If you want to trace intestinal stem cells, you link Cre to a gene that only those stem cells express. Adding a further layer of control, researchers created inducible versions where Cre only becomes active after the animal is given a drug, typically tamoxifen. This “pulse-and-chase” design lets you mark cells at a precise moment and then follow their progeny over days, weeks, or months.2PubMed. Identifying Adult Stem Cells Using Cre-Mediated Lineage Tracing The tamoxifen-inducible approach has been validated across many experimental settings as a reliable tool for gene-level control in stem cell studies.3PubMed. Tamoxifen-inducible Cre-mediated calreticulin excision to study mouse embryonic stem cell differentiation
The system is not without quirks. Cre can sometimes become active in unintended cells, particularly if it is transiently expressed during embryonic development or in the germline. These “leaky” events can label cells that were never supposed to be traced, quietly contaminating the results. Researchers have learned to include careful controls, but unexpected Cre expression remains a well-known pitfall that often goes undetected.4Cell Press. Preventing pitfalls in Cre–lox mouse experiments – Section: Highlights
Multicolor Reporters and Clonal Resolution
A single-color reporter tells you whether a cell came from a marked progenitor, but it cannot distinguish one clone from another. If two nearby stem cells are both marked, their fluorescent descendants mix together and you cannot tell which cells belong to which lineage. To solve this, researchers developed multicolor reporter systems in which Cre activation randomly selects one of several fluorescent proteins, each a different color. A cell might turn green, yellow, red, or cyan, and all of its progeny inherit that same color. Neighboring clones end up in different colors, making them visually distinguishable under a microscope.5PubMed Central. A Method for Lineage Tracing of Corneal Cells Using Multi-color Fluorescent Reporter Mice
These multicolor tools proved transformative for studying tissues where many stem cells sit in close quarters. In the intestine, for instance, multicolor tracing revealed that individual crypts, the tiny pockets where gut stem cells reside, gradually become dominated by a single clone over a period of one to six months. This “drift toward clonality” does not happen because one stem cell is inherently superior. Instead, the stem cells compete neutrally: each cell divides symmetrically, and whether a given lineage expands or vanishes is largely a matter of chance.6PubMed. Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells Independent work confirmed this neutral drift model, showing that the rate of stem cell replacement closely matches the cell division rate, meaning symmetric divisions and random replacement are fundamental to how the gut maintains itself.7PubMed. Intestinal stem cell replacement follows a pattern of neutral drift
CRISPR-Based Barcoding
Multicolor reporters give you a handful of distinguishable labels. For tracing thousands or millions of lineages simultaneously, that is nowhere near enough. Starting around 2016, researchers began harnessing CRISPR genome editing to create a far richer set of barcodes. The approach works by placing an array of CRISPR target sites into the genome and allowing the Cas9 enzyme to introduce small mutations at those sites as cells divide. Each cell accumulates a slightly different pattern of mutations, creating a unique barcode that is passed to its daughters. Because related cells share the mutations acquired before they diverged, computational methods can reconstruct the branching history of an entire organism from these barcodes.
Early demonstrations in zebrafish were striking. By sampling hundreds of thousands of cells from individual fish, researchers found that most cells in adult organs descended from a surprisingly small number of embryonic progenitors.8PubMed Central. Whole-organism lineage tracing by combinatorial and cumulative genome editing Complementary work combined CRISPR-induced barcodes with single-cell gene expression profiling, allowing researchers to simultaneously determine what a cell is (its type and gene activity) and where it came from (its lineage). This dual readout was used to reconstruct developmental lineage trees in zebrafish larvae and in the heart, liver, pancreas, and brain of adult fish.9PubMed Central. Simultaneous lineage tracing and cell-type identification using CRISPR-Cas9-induced genetic scars
These advances dovetail with single-cell sequencing technologies that have matured rapidly. Sequencing-based lineage methods now allow researchers to overlay clonal relationships onto detailed molecular profiles of individual cells, connecting a cell’s ancestry to its current identity and behavior.10Nature Reviews Genetics. Lineage tracing meets single-cell omics: opportunities and challenges
What Lineage Tracing Reveals About Organ Maintenance
One of the biggest practical payoffs of lineage tracing has been settling long-running debates about how adult organs replenish themselves. The liver, for instance, was widely assumed to rely on a dedicated stem cell population to regenerate after injury, much as the bone marrow uses blood stem cells. Lineage tracing overturned that idea. By labeling existing mature liver cells and then tracking their descendants after injury, researchers found that virtually all new liver cells come from the self-replication of existing hepatocytes, not from stem cell differentiation.11PubMed Central. Genetic tracing of hepatocytes in liver homeostasis, injury, and regeneration
The picture is not entirely black-and-white, though. A subpopulation of hepatocytes that express a gene called SOX9 turned out to be bipotent under certain injury conditions: single SOX9-positive hepatocytes could give rise to both new hepatocytes and biliary epithelial cells, the cells that line bile ducts. Researchers confirmed this by tracing individual clones in multicolor reporter mice and finding mixed-lineage clones containing both cell types, each sharing a single fluorescent color.12Stem Cell Reports. Lineage Tracing Reveals the Bipotency of SOX9+ Hepatocytes during Liver Regeneration The lesson here is a good example of what lineage tracing does best: it takes a question like “where do new cells come from?” and provides a definitive answer grounded in direct observation of cellular behavior over time.
Tracking Cancer’s Family Trees
Tumors are not uniform masses. They contain genetically and behaviorally diverse populations of cancer cells, and understanding which subpopulations drive growth, resist treatment, or seed metastases is one of the central challenges in oncology. Lineage tracing has become a key tool for dissecting this heterogeneity. By barcoding cancer cells before they are transplanted or treated, researchers can later ask which lineages expanded, which shrank, and which survived therapy.
In triple-negative breast cancer, for example, single-cell lineage tracing revealed that resistance to an anti-EGFR drug was not acquired through new mutations during treatment. Instead, a rare pre-existing subpopulation of cells was already primed for resistance before the drug was ever applied. These cells expressed elevated levels of a protein called IGFBP2, and overexpressing IGFBP2 was sufficient to make breast cancer cells tolerant to the drug by activating an alternative signaling pathway.13PubMed Central. Single cell lineage tracing reveals clonal dynamics of anti-EGFR therapy resistance in triple negative breast cancer
A broader multi-omic tracing study added another dimension. Researchers found that clones primed for tumor initiation displayed two distinct transcriptional states even before the tumor formed, and these states shared a distinctive pattern of open chromatin. Drug-tolerant clones partially overlapped with tumor-initiating ones but followed their own separate evolutionary paths involving different genetic and transcriptional changes.14Nature Communications. Multi-omic lineage tracing predicts the transcriptional, epigenetic and genetic determinants of cancer evolution In melanoma metastasis, lineage tracing showed that a subset of primary tumor lineages was consistently enriched across multiple metastatic sites, supporting a model of polyclonal seeding followed by selective expansion of the most metastatic subpopulations.15PubMed Central. Single-cell lineage tracing maps clonal and transcriptional dynamics in melanoma metastasis
The recurring theme across these cancer studies is that tumor fate is often pre-encoded. The cells that will resist therapy or colonize distant organs frequently carry molecular signatures that exist before treatment or metastasis begins. Lineage tracing is what makes that conclusion possible, because without knowing which cells are related, you cannot distinguish pre-existing traits from newly acquired ones.
Reading Natural Mutations in Human Cells
All of the engineered approaches described so far require genetic modification, which works in laboratory animals and cell cultures but cannot be applied to living human beings. To trace lineages in human tissues, researchers exploit something that happens naturally: somatic mutations. Every time a human cell divides, it acquires a small number of random mutations. Most of these are harmless passengers, but they serve as natural barcodes. Cells that share the same mutation descended from a common ancestor, and the more mutations two cells share, the more recently they diverged.
Advances in single-cell whole-genome sequencing have made it feasible to detect these mutations at single-cell resolution, opening up lineage tracing in human development and disease.16PubMed Central. Single-cell lineage tracing with endogenous markers In one study of human fetal blood-forming cells, researchers sequenced individual cell colonies and found that a large fraction of mutations were shared between multiple colonies, indicating common ancestry. At eight weeks of development, cells carried an average of about 25 mutations, and by 18 weeks that number had risen to roughly 43, reflecting continued mutation accumulation as cells kept dividing.17bioRxiv. Lineage tracing of human embryonic development and foetal haematopoiesis through somatic mutations By analyzing the patterns of shared and unique mutations, researchers reconstructed branching phylogenetic trees of human fetal development without any genetic engineering.
This retrospective approach also has applications in cancer. Because tumor cells accumulate mutations during clonal expansion, the same phylogenetic logic can reconstruct how a tumor evolved, when key mutations arose, and which branches gave rise to metastases. It is one of the few ways to study lineage dynamics in human patients directly.
Adding Spatial Context
Traditional lineage tracing tells you what a cell’s descendants became, but to do so, you usually have to dissociate the tissue into single cells, destroying the spatial information. A wave of recent technologies aims to fix this by combining lineage recording with spatial transcriptomics, methods that measure gene expression while preserving the physical location of each cell within a tissue.
Several platforms have emerged in quick succession. SPACE-seq integrates CRISPR-based lineage barcodes with spatial transcriptomics to resolve lineage, cell state, and tissue architecture at near-cellular resolution in intact tissue.18PubMed Central. SPACE-seq integrates spatial transcriptomics and lineage tracing in native tissues PEtracer uses prime editing rather than standard CRISPR cutting to generate evolving barcodes that are compatible with both sequencing and imaging, allowing researchers to read out lineage information from tissue sections under a microscope.19PubMed Central. High-resolution spatial mapping of cell state and lineage dynamics in vivo with PEtracer Spatio-DARLIN takes a complementary approach in mice, combining a high-diversity lineage-tracing mouse model with spatial transcriptomics and recovering lineage information from roughly a quarter to half of cells in tissues such as the intestine and brain.20Nature Methods. Spatio-DARLIN enables robust and efficient in situ lineage tracing in mice at single-cell resolution
What spatial lineage tracing enables is answering a new class of questions. Not just “what did this cell become?” but “did related cells end up near each other, or did they migrate far apart? Did neighboring cells in the adult tissue actually come from the same progenitor, or did they arrive from completely different lineages?” These questions matter for understanding how organs organize themselves and how tumors spread within a tissue.
Synthetic Recorders and Computational Reconstruction
Beyond tracing lineage alone, some researchers have built synthetic recording systems that let cells log additional information into their DNA. MEMOIR, for instance, uses a set of barcoded “scratchpads” in the genome that can be irreversibly altered by CRISPR editing. The state of each scratchpad can later be read out in individual cells in situ using fluorescent RNA probes, without needing to extract DNA.21Nature. Synthetic recording and in situ readout of lineage information in single cells Another system, CHYRON, works by repeatedly inserting random nucleotides at a single genomic site in temporal order, creating a molecular tape recorder that captures both lineage and the sequence of events a cell experienced.22Nature Chemical Biology. Lineage tracing and analog recording in mammalian cells by single-site DNA writing
All of these recording approaches generate enormous amounts of data, and interpreting that data requires sophisticated computational tools. Algorithms like Scuphr use probabilistic models to reconstruct cell lineage trees from single-cell DNA sequencing data, accounting for the amplification errors that inevitably arise when copying tiny amounts of DNA from individual cells.23PLOS Computational Biology. Scuphr: A probabilistic framework for cell lineage tree reconstruction Other methods like LinRace integrate barcode data with gene expression profiles, using models of asymmetric cell division to jointly infer both the lineage tree and the molecular states of ancestral cells that no longer exist.24Nature Communications. LinRace: cell division history reconstruction of single cells using paired lineage barcode and gene expression data The computational side of lineage tracing has become as important as the experimental side, because the raw data from barcoding experiments is noisy and incomplete. Without good algorithms, the trees you reconstruct may not reflect what actually happened in the animal.
Lineage Tracing in Plants
Lineage tracing is most commonly associated with animal biology, but plants pose their own fascinating questions about how cells contribute to growth and regeneration. Plants grow from organized clusters of stem cells called meristems, and understanding how individual meristem cells give rise to different tissue types has practical implications for agriculture and plant breeding.
A Cre-lox lineage tracing system adapted for plants was used to trace the behavior of quiescent center cells in the root, a small group of rarely dividing cells thought to serve as an organizing hub for root stem cells. Under normal conditions, these cells stayed put and did not divide to replenish surrounding stem cells. But when stressed with osmotic pressure, they began dividing and contributed progeny to the vascular, cortex, and endodermis cell layers, roles they do not normally play. The same system was applied to shoot regeneration, revealing that the shoot apical meristem arises not from a single progenitor cell but from a group of shoot progenitor cells, a conclusion drawn from the mosaic labeling patterns that resulted from transient Cre activation.25PubMed Central. Tracing meristem cell lineages during plant development and regeneration by the all-in-one CRE/LOX system
Tracing Immune Cells by Their Natural Barcodes
The adaptive immune system provides a unique natural experiment in lineage tracing. Each T cell carries a unique T cell receptor generated by random DNA rearrangement. When a T cell recognizes a threat and proliferates, all of its clonal descendants share that same receptor sequence, creating a natural barcode that can be read out through sequencing. Paired single-cell RNA and T cell receptor sequencing now enables researchers to profile T cells in terms of both their clonal lineage and their functional state simultaneously.
This approach has been applied to study how T cells behave in and around tumors. By tracking T cell clonotypes across tissues, researchers found that certain helper and exhausted killer T cells in the tumor were clonally linked to progenitor cells in tumor-draining lymph nodes. The data also revealed progressive exhaustion trajectories: T cells closer to the tumor microenvironment showed more advanced signs of functional exhaustion than their clonal relatives in nearby lymph nodes.26Cancer Cell. Paired single-cell RNA and T cell receptor sequencing of T cells in the tumor microenvironment – Section: Summary This kind of clonal tracking is reshaping how immunologists think about anti-tumor immunity, because it connects the cells actively fighting a tumor to the reservoir populations that feed them.