Single-Cell Proteomics: Key Takeaways from SCP2021

In this blog, Fios bioinformatician Katerina Boufea gives an overview of the key takeaways from the fourth Single-Cell Proteomics Conference. SCP2021 took place from the 16th-18th August, both in Boston, USA, and virtually.

SCP2021 aims to be a platform for sharing the know-how of single-cell proteomics and making it widely accessible. With this in mind, the event included different workshops and talks on topics ranging from sample preparation to data analysis.

Different speakers presented various platforms and protocols, along with their advantages and technical challenges. Some of the most interesting technological advancements included the automated microdissection of cells. This new process makes single-cell proteomics experiments less time-consuming and improves the consistency of sample preparation. This can help to better scale and incorporate single-cell proteomics into the clinical practice for both diagnostic and prognostic purposes.

There is currently no need for increasing the number of detected peptide features. However, many opportunities exist for developing strategies for data acquisition and sequencing; all this to reliably determine the sequence of the peptide features and increase accuracy in quantifying protein abundance.

Single-Cell Proteomics and Bioinformatics

In the field of bioinformatics, various presentations discussed new methods and software for analysis of single-cell proteomics data, such as the MaxQuant.live and MaxDIA additions in the MaxQuant suite, SCP and IceR. Researchers can use these tools for data generated from different platforms and for various data analysis steps, from protein spectrum identification to interpretation. While there is overlap in their use, there is not a single method or tool that fits all single-cell proteomics data. Depending on the technology, different computational methods are more suitable to analyse the data taking into account platform-specific technical bias. Missing values and batch effects between samples are common issues similar to single-cell RNA-sequencing data. One of the main open challenges in single-cell proteomics is to characterise the technical noise in the data;  so that it can be modelled during downstream analyses, similar to what has been done for single-cell RNA-sequencing data.

Applications

A wide range of applications were presented. Single-cell proteomics studies have shown greater than expected cell to cell variability which indicates that we need to study protein abundance within cell populations. Additionally, there is a need to study cell heterogeneity in space and time. Other interesting applications included studies of nuclear proteins and how these control gene expression and studies of transcription factors in a cell type-specific manner. There were also studies of post-translational modifications in neurons, as well as studies on embryonic development.

Various studies integrated single-cell proteomics with other single-cell omics data, such as metabolomics, RNA-sequencing, spatial information, DNA-sequencing, and chromatin accessibility data. This is a recurring pattern we also see in various recent publications and multi-omics can achieve a more complete understanding of cellular biology. Researchers have made tremendous progress during the last five years; however, this is still the beginning and more advances are expected to change:

• the quality of the data we obtain,
• the information we can extract from the data,
• the technical bias we will have to account for during analysis.

See also:

SCP2021

Proteomics Data Analysis

Publications featuring proteomics analysis from Fios Genomics