Automation of Liquid Handling for Long-Read Library Preparation

SPT Labtech’s technology for automated library construction enabled large-scale whole genome sequencing for the Darwin Tree of Life Project

By Maryia Karpiyevich, PhD

Credit: Richard Jones/Science Photo Library/Getty Images

Whole genome sequencing (WGS) is a comprehensive method that enables scientists to assess the genetic blueprint of diverse species and observe critical inter and intraspecies variations. WGS data allows the determination of critical factors in eukaryotic evolution and survival, aiding biological, biotechnological, and conservation research.

The Wellcome Sanger Institute’s Darwin Tree of Life Project is one of several flagship initiatives aimed at advancing our understanding of genetic biodiversity. The Project aims to utilize WGS approaches to generate over 70,000 high-quality reference genomes covering all known species of plants, animals, fungi, and protists that live within the bounds of the British Isles. By exploring the genetic profiles of these diverse species, the Darwin Tree of Life Project aims to generate reliable and robust open-source data that can be used for ongoing study.

The process of sequencing, assembling, and publishing such a vast quantity of genomic data requires lengthy and laborious liquid handling workflows, which can create bottlenecks. Manual liquid handling also carries an increased risk of human error and ergonomic strain. Plus, variability between users introduces additional confounding factors into experimental protocols.

Automated liquid handling gives the agility to flexibly scale with demand while also offering standardization of DNA library preparation workflows. However, the integration of this technology raises some challenges due to the sensitivities of long-read protocols. Therefore, it is essential that due care is taken to maintain DNA quality and workflow integrity.

For successful long-read data acquisition, automated protocols must cause minimal DNA disruption and preserve sample integrity. It is essential to avoid excessive DNA fragmentation as smaller fragments can interfere with long-read sequencing processes.

Some DNA library preparation reagents are highly viscous, posing challenges when dispensing at low volumes using traditional air-displacement technologies. Poor pipetting control can lead to inconsistent results and can compromise on the coverage of sequence reads.

Although manual sample preparation methods offer flexibility and can be carried out across multiple benches, automated solutions need to fit firmly within a dedicated space. Typically, such spaces are constrained due to the necessity of other instruments. Additionally, operating software should be user-friendly and require minimal training to facilitate swift adoption by all relevant users.

Manual and automated workflows were both tested using the PacBio SMRTbell long-read prep kit (SPK3) for sequencing with PacBio’s Revio platform. Automated workflows were performed using SPT Labtech’s all-in-one platform for genomics, fireflyR (Figure 1). Its high-throughput capability is facilitated by 96- and 384-channel air-displacement pipetting functionality, and tip arrangement in collared strip columns ensures consistent tip pickup and pipetting accuracy.

Moreover, the firefly platform’s ability to handle viscous liquids accurately through a positive-displacement dispense head and configurable parameters enabled easy and efficient integration with the pipeline. The platform’s compact footprint and intuitive visual software provided easy accessibility. Also, application scientists at SPT Labtech were available to support and advise on method development and validation.

Automation protocols were based on the published SPK3 workflow; however, liquid handling steps were modified to facilitate high-throughput library preparation on firefly. Master mixes were prepared manually to accommodate the volumes of reagents provided in the kits and dispensed using the firefly platform’s positive-displacement dispense head.

Optimizing pipetting and dispensing parameters was critical for accurate pipetting and uniform dispensing across all samples, as well as ensuring DNA integrity was not compromised. Additionally, firefly’s noncontact dispense technology reduced

plastic tip use, contributing to cost efficiency and decreasing waste. Plates were removed from the firefly deck manually for incubation in standard thermocyclers, and safe stopping points were established to ensure quality control and allow the process to be performed over multiple days if necessary.

With the methods developed, the Wellcome Sanger Institute aimed to compare both approaches to explore whether manual and automated systems could deliver results of equivalent and sufficient quality. Validation testing covered diverse genomes due to the nature of the Darwin Tree of Life Project.

Eight species were selected: Adalia bipunctata (ladybird), Dunaliella primolecta (green algae), Metschnikowia zobellii (marine yeast), Mus musculus (mouse), Physella acuta (freshwater snail), Quercus robur (English oak), Teleogryllus oceanicus (cricket), and Homo sapiens (human cell line).

Manual and firefly-automated library sets contained triplicates of the eight species. To produce a more informative comparison, variables (such as master mixes) were controlled and split between manual and automated methods. Both methods were compared by measuring library yield and size. Library size was validated by measuring nucleic acid traces in each sample using the Agilent Femto Pulse. Checkerboard validation of SPK3 workflow on firefly was performed with cricket library DNA as a positive control and elution buffer as a negative control to ensure uniformity and check for intraplate contamination (Figure 2).

Manual sample preparation in line with the manufacturer’s guidelines is regarded as the gold standard method for long-read library preparation. The Wellcome Sanger Institute’s results showed comparable library yields (as percentage of input) and coefficients of variability between the manual and automated methods (Figure 3A), validating the automated method as a suitable replacement for manual workflows. Moreover, trace analysis of nucleic acid content in samples prepared by automated and manual methods delivered similar results, validating the size of the libraries created (Figure 3B).

Although long-read sequencing data provides a wealth of information about species diversity, lengthy and highly variable library preparation protocols can arise as a productivity bottleneck, with additional risks of ergonomic strain. Automation provides an obvious solution to the challenges of manual liquid handling workflows. However, to ensure successful adoption, considerations must be taken to select a technology that meets the functionality and accessibility requirements of the laboratory.

firefly proved both accessible and versatile, delivering results that were of equal quality to those achieved manually (Table 1), but at a faster output and with less of a burden on valuable researcher time, allowing the lab to operate more efficiently. As a result of automated SPK3 workflows on firefly, high-quality libraries were obtained from a variety of species, including the notoriously difficult algae and snail species.

This Project with the Wellcome Sanger Institute demonstrated that firefly is a viable option for automated long-read library preparation by providing reliable DNA library generation from a variety of biological organisms.

Maryia Karpiyevich, PhD, is a product development scientist at SPT Labtech.