Modern Plant Genomics: Sequencing and Cloud-Driven Workflows

15 Dec 2025

WRITTEN BY

Suhasini Singh

What is driving the shift from microarray to NGS?

Since the early days of plant genomics, microarrays have been the standard tool for measuring gene expression. Their affordability and practicality made them widely adopted, and they remain useful in some settings. As sequencing technologies have advanced, however, certain limitations have become clear. Microarrays rely on known probes, which restricts their ability to detect new transcripts or unexpected genetic variation. They also have a limited dynamic range, so subtle differences in expression may not be captured.

High throughput sequencing is gradually becoming the preferred approach as it overcomes many of the constraints seen with microarrays. By reading sequences directly, it can reveal new genes, isoforms, structural variants, and single nucleotide changes that would otherwise go undetected. As sequencing costs fall and workflows become more manageable, more laboratories are able to bring sequencing into their routine studies.

Why Cloud-based Analysis is Becoming Essential

Modern sequencing produces datasets that are often too large for local machines to handle. Large genome assemblies, long read datasets, and pan-genome projects require storage and compute capacity that many labs cannot support on site. As a result, cloud platforms have become a practical way to run heavy workflows, share data across research groups, and scale analyses.

Supporting Technologies: Long Reads and Pan-Genomes

Many horticultural species have complex genomes shaped by hybridization and domestication. Often, this complexity cannot be resolved by short reads alone, but in combination with long reads, we get stronger, chromosome level assemblies that improve annotation, trait mapping, and variant discovery.

A single reference genome cannot capture the full diversity within a species. Some genes are shared across all varieties, while others appear only in certain lines and influence traits such as stress tolerance, yield, or local adaptation. Pan-genomes capture this broader range of variation and help identify traits that would be missed with a single reference. These studies depend on large sample sets and substantial compute, which suits cloud based analysis.

This approach also benefits orphan and underused crops. Many regions rely on local species with limited genomic resources, and new sequencing methods now allow them to be studied with the same depth as major crops.

What These Advances Enable

Better trait discovery: Improved assemblies and more accurate variant calls lead to stronger genome-wide association studies, and better inputs for genomic selection. These insights help guide breeding for resilience and productivity.
Integration of different data layers: Modern studies often combine DNA sequences, RNA expression, epigenetic patterns, and metabolite profiles. These integrated datasets give a more complete view of how traits arise. They also support precise gene editing with tools such as CRISPR, base editing, and prime editing.

How Strand Supports Plant Genomics Research

Our team has carried out genomics and transcriptomics studies in more than sixteen plant species over the past three years. These include fruits, vegetables, crops, millets, and even insectivorous plants. We work with newly assembled genomes, multi-omics studies, and complex datasets. This includes one of the largest plant genomes, onion, with an estimated size of 16 Gb.

We use both open source tools and our dedicated platform, Strand NGS. The platform supports DNA, RNA, and microbial sequencing data and helps run complete workflows from raw data to analysis and visualization.

RNA-Seq Analysis Phase	Strand NGS Capacities	Application in Plant Genomics
Pre-Processing & Alignment	Read/Sample filters, Align reads, Discard poorly aligned reads, Check Gene Body Coverage	Handles alignment for large genomes; crucial for managing samples with high novel events.
Variant and splicing analysis	SNP analysis - detection and advanced filters to get high confidence SNPs, Splicing analysis and Gene Fusion detection	Help uncover genetic and regulatory changes that drive key plant traits, including stress tolerance, development, and overall crop performance.
Quantification & Normalization	Quantify Gene Expression, Normalize Gene Expression, Filter genes expressed at low levels	Ensures accurate quantification for large DE comparison sets (e.g., see DEG comparison in the foxtail millet study webinar)
Advanced Statistics	Statistical tests, Confounding Variable Analysis, Correlation, Class Prediction	Facilitates complex cross-tissue (Leaf vs. Root) and cross-genotype comparisons.
Downstream Analysis, Visualization & Reporting	PCA, t-SNE, UMAP, Cluster entities/samples, volcano plots, violin plots, Gene Ontology, network and pathway analysis	Generates visually compelling and biologically informative outputs (e.g., see PCA and correlation plots in the foxtail millet study webinar).

Working With Difficult Genomes

A key strength of Strand NGS is its ability to handle complex alignment cases that often cause open-source tools to fail. For samples with a high frequency of novel events, such as onion, the platform uses a custom, scalable solution that splits, aligns, and merges reads to overcome memory and thread limitations and maintain data completeness. Ongoing updates are strengthening native support for high-novelty samples, making the platform increasingly robust for difficult plant genomes.

Our Workflow

We have built supporting infrastructure that allows these analyses to run reliably at scale. The system is designed to handle complex datasets, automate key steps, and keep variant curation clear and traceable, ensuring that results can be used confidently for downstream decision-making.