Technical variability can have a significant impact on single-cell experiments, potentially affecting the accuracy and reproducibility of the results. Unlike biological variability, which reflects true differences in gene expression between individual cells, technical variability arises from experimental and processing inconsistencies. In this feature, we explore the impacts of technical variability, and how standardisation might be the way forward. With plenty of input from experts in the single-cell space, we hope this feature helps you with your experiments.
How can technical variability impact an experiment?
Variability in single-cell experiments can stem from several sources, including cell isolation, RNA capture efficiency, sequencing depth and data preprocessing.
High technical variability in single-cell experiments affects data quality, interpretation and reliability, potentially masking true biological signals. Here are just some of the impacts this can have on your work:
Dropout Events
- Dropouts occur when genes are undetected in certain cells due to low capture efficiency or sequencing depth.
- This can create false impressions of selective gene expression in certain cell populations due to technical noise.
- Results in potential misclustering, where similar cells may appear in different clusters or different cells in the same cluster.
Challenges in Identifying Cell Subtypes
- Misleading clustering can complicate the identification of true cell subtypes, states or rare populations, leading to inaccurate conclusions about cellular diversity.
Batch Effects
- Systematic inconsistencies between batches (batch effects) can falsely suggest biological differences.
- For instance, samples processed in different batches may appear to show treatment effects that are actually batch-related artifacts.
Gene Expression Quantification
- Technical factors like PCR biases and sequencing depth can lead to over- or underrepresentation of certain genes.
- This skews gene expression profiles, potentially affecting the identification of key biological pathways or significant genes in downstream analyses.
These technical issues can hinder the validity and applicability of findings derived from single-cell experiments, impacting the robustness of biological conclusions.
Standardisation: The way forward?
“Having a standard operating procedure makes achieving reproducibility easier. Success is a lot of small things done right.” – Luciano Martelotto (Associate Professor, Head of the Development Laboratory, Adelaide Centre for Epigenetics (ACE), South Australia ImmunoGENomics Cancer Institute (SAIGENCI)), speaking in our recent Single-Cell and Spatial Playbook.
To address these challenges, many researchers and organisations have called for the implementation of standard practices. For example, establishing community guidelines for best practices in experimental design, data collection and analysis can provide a foundation for more standardised approaches. Consortia like the Human Cell Atlas and initiatives led by the National Institutes of Health (NIH) are already working toward such goals by recommending protocols and creating reference datasets. Secondly, developing open-source, reproducible software tools with transparent documentation can help ensure that data analysis pipelines are accessible and can be consistently applied across studies. Encouraging the adoption of FAIR principles (Findability, Accessibility, Interoperability, and Reusability) for single-cell data is also essential to promote data sharing and integration.
Finally, investing in benchmarking studies that systematically compare different methods and protocols under similar conditions can provide the researchers with much-needed performance data on various approaches. These benchmarks can guide researchers in selecting methods that minimize technical variability and maximize reproducibility. By prioritising these initiatives, more standardised and robust frameworks can be developed, facilitating reliable discoveries and paving the way for the translation of single-cell insights into real-world applications.
At our recent FOG Live: Single Cell and Spatial Event, we asked our audience to what extent they agree with this statement: “A standard operating procedure would help researchers standardise single-cell and spatial data analysis”. The majority of attendees agreed that standard operating procedures would be a helpful step forward.
The Human Cell Atlas: Standardising Single-Cell for Real-World Applications
The Human Cell Atlas (HCA) initiative has been instrumental in promoting standardised protocols for single-cell experiments, with the goal of creating a comprehensive map of all human cell types. Given the complexity and variability inherent in single-cell technologies, scientists involved in the HCA recognised early on that standardised methods were essential for generating high-quality, reproducible and comparable datasets across the global research community. Here are just some of the ways the HCA is striving to improve standards.
Protocol Standardisation and Documentation
The HCA has developed and recommended detailed protocols for each stage of the single-cell experimental process, including sample collection, cell dissociation, RNA extraction, library preparation and sequencing. By creating standardised, widely accessible protocols, the HCA aims to minimise technical variability and ensure that all participating labs are able to generate data that is directly comparable. These protocols are publicly available and accompanied by clear documentation, making it easier for researchers to follow a consistent process.
Creation of Reference Datasets
To set benchmarks for quality and reproducibility, the HCA has established ‘reference datasets’ for various tissue types and cell populations. These datasets serve as gold standards that other researchers can use to compare their own data, helping to ensure that their single-cell workflows are producing results that align with community standards. These reference datasets are generated using the HCA’s standardised protocols and are often the product of collaborative efforts among multiple labs, which further reinforces protocol consistency and reliability.
Development of Quality Control Standards
The HCA has introduced stringent quality control (QC) standards to evaluate sample preparation, sequencing quality and data integrity. These QC guidelines help researchers assess critical aspects such as cell viability, RNA integrity and sequencing depth, all of which are essential for obtaining accurate and reliable single-cell data. By standardising QC measures, this helps reduce the risk of batch effects and technical artifacts, promoting consistency across studies.
This feature was written using intelligence gathered from the single-cell community at our recent FOG Live: Single-Cell and Spatial event, The Festival of Genomics and Biodata, the Single-Cell and Spatial Playbook 2024 and conversations with experts. With thanks to Aridaman Pandit (Senior Research Scientist, Abbvie), Luciano Martoletto (Associate Professor, Head of the Development Laboratory, Adelaide Centre for Epigenetics (ACE),South Australia ImmunoGENomics Cancer Institute (SAIGENCI)), Tancredo Massimo Pentimalli (PhD Candidate, Max Delbruck Center), Ashleigh Lister (Senior Research Assistant, Earlham Institute), Nancy Zhang (Professor of Statistics, University of Pennsylvania), Kristen Beaumont (Icahn School of Medicine), Adam Cribbs (Group Leader, University of Oxford) and Mathew Chamberlain (Principal Computational Scientist, Johnson & Johnson Innovation).
