Trajectory Analysis-1: Concepts & Applications

Duration: 40 Minutes

Lecture Material:

Quick Summary


In the last decade, significant advancements have been made in the realm of cellular biology. One such innovation, single-cell RNA sequencing (scRNA-Seq), has offered the scientific community a deeper look into individual cell gene expressions.

Distinguishing Between Cell States and Cell Types

It is imperative to understand the nuanced difference between cell states and cell types. While cells continuously undergo alterations and adapt to various stimuli, these transitory configurations are termed 'cell states'. On the other hand, 'cell types' represent a more stable classification characterized by consistent histology, specific functionalities, and an unvarying transcriptomic signature.

Analyzing the Transcriptomic State Space

The vast amount of data yielded by scRNA-Seq, which spans thousands of genes, presents its set of challenges. Each cell's gene expression can be visualized in a transcriptomic state space. This space is critical for understanding cellular responses and differentiation processes.

To navigate this vast data, dimensionality reduction techniques such as PCA, UMAP, t-SNE, Diffusion Maps, and PHATE are utilized. These methods simplify the data, enabling more precise identification of cell states and types.

Pseudotime: A Theoretical Interpretation of Cellular Progression

Pseudotime is a concept used to understand the sequence of cellular development. It is not a real-time measure but an estimation based on changing gene expressions. This theoretical framework becomes particularly essential when studying complex processes like embryogenesis or hematopoiesis, revealing the stages of cellular transitions.

The Importance of Root Cells in Trajectories

In the context of cellular development, root cells signify the inception. Determining the trajectory's initial cell is fundamental to deciphering the sequence of cellular progression. Various methodologies are employed to identify root cells, such as entropy-based techniques and the RNA-velocity method.

Existing Limitations in Transcriptomic Analysis

Our comprehension of cell states and transcriptomic dynamics is still evolving, and there are several challenges to consider:

  1. Sampling Bias: It is assumed that the sampled cells represent diverse stages of a biological process. Deviations from this can lead to inaccurate conclusions.

  2. Interconnected Regulatory Mechanisms: The concurrent occurrence of cell cycle regulation with differentiation introduces complexity into trajectory analyses.

  3. Dependence on Gene Expression: Relying solely on transcriptomic data may not provide a holistic understanding of the cell's state. Factors such as the cell’s chromatin state or DNA methylation patterns are equally pertinent.


The quest to unravel the mysteries of cellular differentiation and development is ongoing. The introduction of technologies like scRNA-Seq and methodologies for trajectory analysis are pushing the boundaries of our understanding. Though challenges persist, the continuous evolution of these methods holds promise for a deeper insight into cellular biology.