Reflecting on single-cell RNA seq, spatial transcriptomics, their integration to advance fundamental and translational biology, and on how bioinformatics evolve

Through bioinformatics, where data processing and analysis techniques are in continuous evolution as the experimental methods to explore the -omics advance, groundbreaking innovations continue to redefine our understanding of the biological world. New methodologies continuously push the boundaries of what is possible in data analysis, as we have covered in recent articles presenting, for example, a new tool for simple and powerful integration of single-cell RNA data from heterogeneous sources, some new methods to accelerate proteomics analyses, and our own end-to-end solution for Omics analysis, ONex.

The fast pace at which the bioinformatician’s (and hence the biologist’s) toolkit evolves sets the stage for today’s blog post. Here we will explore how new tools continuously come out that help researchers to get the most out of two of the current stars in biology and precision medicine: single-cell RNA seq and spatial transcriptomics. After presenting some modern examples of the application of these powerful tools to precision medicine and biology, we then touch on how we at Nexco deal with the fast pace at which experimental techniques and their associated bioinformatics solutions flood literature.

Single-Cell RNA seq and Spatial Transcriptomics

Much of the focus of our work, as reflected in our blog too, involves some of the most powerful experimental tools to probe cells, tissues, and their evolution over time and space: Single-Cell RNA sequencing, Spatial Transcriptomics, and their blend.

We covered already here the BRAIN Initiative Cell Census Network, which combined single-cell and spatial transcriptomics to catalog the type and location of cells present across all parts of the mouse brain in unprecedented detail. Now, we thought this was a good moment to set the spotlight on a few more works that further demonstrate the power of these two techniques, specifically regarding applications to fundamental biology and translational medicine.

Predictive Power in Personalized Medicine

In personalized medicine, the quest for targeted therapies takes a leap forward with the aid of predictive analytics. An interesting modern example, this work presents PredicTCR, a machine learning classifier that processes scRNA-seq data to identify tumor-reactive T cell receptors swiftly and accurately, certainly faster and with much better confidence than alternative tools. By deciphering the genetic signatures of tumor-reactive TILs, PredicTCR brings medicine one step closer to a future of personalized transgenic T cell therapies that can be optimized promptly and without delays, giving more hope to cancer patients.

From Anatomy to 3D Molecular Understanding

Integration of spatial and single-cell transcriptomic data can take our whole understanding of biology to new levels, especially when it’s about developmental or differentiation programs as measured by tracking specific patterns of RNA expression.

For example, integration of spatial and single-cell transcriptomic data unveiled in this work some quite complex insights into organogenesis. Through the application of image-based single-cell transcriptomics using sequential fluorescence in situ hybridization, the mRNA expression level was detected for hundreds of target genes in tissue sections of mouse embryos. By integrating spatial context with multiplexed transcriptional measurements and utilizing two single-cell transcriptome atlases, the study could characterize cell types across the embryo. This revealed spatially resolved gene expression patterns in the form of a high-resolution 3D map that facilitates the characterization of fundamental steps in tissue patterning and in cell fate decisions during organogenesis, thus shedding new light on the molecular mechanisms underlying developmental processes. In particular, the spatial dimension allowed researchers to uncover axes of cell differentiation that are not apparent from scRNA-seq data only.

Decoding Developmental Dynamics

From the earliest stages of embryogenesis to the patterning of complex tissues, development unfolds through a strictly controlled gene expression program that fundamental biology aims to dissect — knowledge of profound impact in medicine. This work presents GEMLI, a computational method available as an R package to trace cellular lineages and unravel the molecular mechanisms driving differentiation in scRNA-seq datasets. As an application case, the very same paper describing the tool shows how on cancer biopsies GEMLI detects previously unknown gene expression changes at the onset of cancer invasiveness, a critical timepoint for therapeutic attack.

How Experiments and Bioinformatics Evolve Continuously, and how we Track This at Nexco

Now fueled by the growing power of scRNA-seq and spatial transcriptomics, but propelled by other technologies in the past and surely yet technologies to come in the future, scientists constantly attempt to squeeze the most out of data. Thus, bioinformatics finds itself unceasingly producing new methods, models, databases and software.

Better statistics, better ways to integrate datasets, more complete atlases, richer visualizations, and developments at various other fronts all pursue to unlock biological insights and to provide medicine with better tools to diagnose and cure disease. From personalized medicine to fundamental research, the arsenal of experimental and computational techniques holds the promise of revolutionizing our understanding of life itself.

We at Nexco are very much aware of this continuous evolution, so we have our experts’ eyes locked on to the literature reporting new experimental and computational methods that zoom into biology. Today, for example, our browsing of very recent literature for tools resulted in PredicTCR and GEMLI covered above, plus a new tool called FAVA that helps to infer high-quality functional association networks from scRNA-seq and proteomics data, and SPANN which facilitates the alignment of spatial transcriptome samples with RNA data prototypes thus helping to merge both kinds of datasets into detailed spatial transcriptomics maps.

As you’ve seen in our blog, routine exploration of recent literature along all of Nexco’s service fronts from bioinformatics and statistics to structural biology and virtual reality applications to them, allows us to stay up to date with all new technologies and thus to serve you with the most modern answers and solutions to your questions and problems.