The emerging role of open technologies for community-based improvement of cryopreservation and quality management for repository development in aquatic species

Abstract

Genetic resources of aquatic species are of tremendous value, but worldwide these are maintained almost exclusively as live populations. This is extremely expensive and insecure, and largely results from a pervasive lack of production capability, quality management, and reproducibility in cryopreservation that are barriers in development of germplasm repositories. Community-based technology approaches are emerging that stimulate research previously limited by a lack of affordable, customizable equipment. Open-access technologies can provide for custom design and fabrication not available through traditional manufacturing. This can assist repository development with robust production methods and strong quality management, and can greatly improve reproducibility and standardization. Open technologies can support establishment of new communities of users, makers, and developers that collectively strive to develop open hardware in a distributed (i.e., non-centralized) fashion that can yield aggregate throughput. This occurs through use of consumer-level tools, supplies, software, and equipment, free exchange of designs and modifications, and a shared sense of mission. For cryopreservation and repository development, we have identified 14 categories of open hardware for a processing pathway, and six categories for a quality management pathway. Open hardware offers economic incentives to develop repositories for aquatic species, something that has not occurred despite 70 years of research largely focused on protocol development rather than practical applications. Advanced development of custom scientific hardware enhancing open-access technologies will be facilitated by interdisciplinary collaboration across biological and engineering fields. This manuscript is a contribution to the Special Issue in memory of Dr. Duane Garner, a leader in the sperm biology.

1. Introduction

Sperm cryopreservation offers many benefits including commercial opportunities such as the multi-billion dollar global markets that have existed for decades for livestock semen, and development of germplasm repositories to maintain, characterize, and protect valuable genetic resources (Purdy et al., 2016; Liu et al., 2019a). These benefits have been largely realized in mammalian applications, but even though there has been 70 years of research, mostly aimed at protocol development, commercial and repository development have only begun to be realized for aquatic species. In response to this, there have been efforts to overcome the largest barriers which include lack of production volume, quality management, and reproducibility. These efforts have included using established technologies such as equipment routinely used for dairy bull semen processing and practices to increase throughput and improve quality control (Paul Lang et al., 2003; Dong et al., 2005).

The model that was implicitly put forward in those efforts was to generate market demand and stimulate repository development as a means to motivate established industrial manufacturers to produce new equipment and supplies that would be appropriate for aquatic species. This was because at that time (1990s to early 2000s) large companies were the only source and purveyors of such technologies. Since then, especially in the past 10 years, there has been a revolution in what we term as “open technologies” that have become available to support an emerging “open hardware” movement (Gibney, 2016). These include consumer-level fabrication technologies (e.g., 3-D printing), open-license computing options (e.g., design software), and electronic, data management, and programming capabilities (e.g., microprocessors). The advent of these new readily accessible and inexpensive technologies has resulted in decentralized production, and has greatly reduced or eliminated the reliance on established companies with industrial approaches to be the sole source of innovation and production of tools, devices, and other hardware necessary to provide solutions for the problems inherent in cryopreservation for aquatic species. Most researchers are familiar with individual technologies such as 3-D printing or microcontroller boards (e.g., Arduino), but few outside of engineering or maker communities recognize the diversity of options and the opportunities available through integration of open technologies.

Open technology approaches occur as a result of actions in vibrant communities that have developed around powerful yet inexpensive consumer-level design and fabrication advances that greatly reduce the entry barriers to custom development of hardware such as scientific devices. This means that existing groups, such as cryopreservation researchers or fish hatcheries with unsatisfied needs, can gain new modalities to address those needs in a progression where there is adoption of open hardware and become user communities of the new modalities. Moving forward, there are opportunities to form communities that can produce distribute, share, modify, and improve hardware with a further impetus being the emergence of hardware development self-sufficiency (Liu et al., 2019a). Overall, the expression of open technologies is based on rapid and continued sharing of advances among those in these communities based on granting of freedoms with a range of protections and licensing agreements that can include existing provisions such as trademarks and patents. These freedoms can include permission to fabricate, modify, or distribute designs created and shared by others, with specified restrictions on use (e.g., with or without commercialization) or provisions for ascribing credit. This thinking can trace its roots to open-source software initiatives (Raymond, 1999; Hippel and Krogh, 2003) and has coalesced into organizations such as Creative Commons (creativecommons.org) which provides a system of licensing and provides structures for placing innovations into the public domain.

Considering the tremendous need to protect, manage, and distribute the genetic resources of aquatic species, it has become increasingly important to identify mechanisms beyond traditional research to produce standardized approaches to gamete cryopreservation. The goal in this article, therefore, is to examine the emerging role that open technologies can play to improve throughput and quality management of cryopreservation for repository development. Objectives are to: 1) provide an overview of previous and emerging approaches to repository development; 2) explain open-access technology concepts; 3) introduce possibilities for transitioning from traditional manufacturing to open technologies; 4) provide examples of open hardware relevant to applied cryopreservation pathways, and 5) examine the participation and community development necessary for utilization of open technologies. The first section below will review the limitations of existing models for repository operation and identify opportunities to addresses these issues with open technologies.

2. Approaches to develop germplasm repositories

2.1. Central facilities, mobile laboratories, and aggregate throughput

The primary model to transform germplasm such as sperm samples into repository accessions is processing through central facilities (e.g., Zebrafish International Resource Center, ZIRC). These facilities would have consistent and adequate funding to support core capabilities, such as advanced equipment, experienced staff, dedicated space, established sample processing and quality management pathways, and high throughput (e.g., hundreds or thousands of samples per day). Currently there are less than five facilities in the United States that have basic central processing capabilities and a longstanding focus on cryopreservation for aquatic species. With this limited number of facilities, it is difficult to gather germplasm from hatcheries and wild populations, and it would cost millions of dollars to build, outfit, and staff new cryopreservation facilities across the United States. Although hundreds of cryopreservation protocols have been published, the ultimate problem is that these protocols can only be reliably conducted in a small number of facilities with specialized personnel and equipment. For example, among the >1,400 laboratories (ZFIN, 2021) worldwide (ZFIN) that use and produce zebrafish lines, only six (less than 1%) have submitted samples to ZIRC during the past 20 years (personal communication, Z. Varga, Director). This situation also exists with other aquatic species, and as such, more than 99% of biomedical researchers in aquatic species lack capabilities in cryopreservation and associated quality management.

An alternative strategy to address these limitations of centralized processing is through development of mobile cryopreservation capabilities (Childress et al., 2018; Childress et al., 2019). With mobile laboratories, samples could be processed and cryopreserved within hours of collection, reducing variability due to effects of refrigerated storage and transportation (Childress et al., 2018). This strategy can provide access to genetic resources and user groups that would be difficult or impossible to accommodate with typical centralized processing, facilitating repository development. Mobile laboratories are most suitable for destinations of < 1,000 miles, however, even though there are these advantages there are requirements of expensive equipment and highly trained personnel.

“Aggregate throughput” is a term used in this article to describe a third strategy to achieve large-scale production that is independent of centralized processing. With this strategy samples are processed by community members themselves, with relatively low throughput individually, yet combining for high throughput when aggregated in a unified repository system (Hu et al., 2017). For example, it would be relatively easy for an individual laboratory to cryopreserve several dozen samples if simple, reliable, and inexpensive options existed for processing, and submission of samples and associated information to a central location. As such, if thousands of laboratories or users could submit high-quality samples to repositories, the aggregate throughput of tens of thousands of accessions would be comparable to that achievable by a central facility.

2.2. Achieving reproducibility of aggregate throughput using open hardware

The key (and greatest problem) in achieving successful aggregate throughput is production of samples with adequate quality. This is challenging in any context because of the existing pervasive lack of reproducibility among cryopreservation research communities for aquatic species (Torres and Tiersch, 2018). The hardware used in cryopreservation is extremely variable, and it is difficult to find any two groups working with aquatic species that conduct the same techniques and utilize the same equipment. Improving the functionality of hardware can greatly facilitate standardization of processes, reporting, and training. The hardware available for aquatic user communities, however, has several major issues including cost (commercial equipment is too expensive for widespread use), reproducibility (self-made devices are often too crude to be standardized), portability (equipment is often too cumbersome or valuable to transport), and customizability (proprietary products, often developed for use with mammals, are difficult or impossible to be customized).

Cooling hardware can be considered as an example. There can be expenditures on commercially available programmable freezers of $15,000 to $50,000 (US), and thus these freezers are not cost-efficient for non-commercial users who have limited budgets, or only occasional cryopreservation needs. Commercially available non-programmable products (e.g., Coy et al., 2019) are budgetarily more feasible (costs being several hundred US dollars). The slow cooling rates (e.g., 1 °C/min) designed for banking of somatic cells or tissues (Yokoyama et al., 2012; Lin et al., 2013), however, are not suitable for conducting sperm cryopreservation procedures, because most animal sperm cells are frozen at more rapid cooling rates (e.g., 5 – 25 °C/min), and fixed rates are often not useful for research where the aims are to develop cryopreservation protocols.

Self-made devices are often constructed due to the prohibitive costs and challenges of equipment utilization associated with field freezing applications. With the majority of these customized devices, there is utilization of the temperature gradient in nitrogen vapor to produce desired cooling rates (Wayman et al., 1997; Purdy, 2006; Clulow et al., 2008). This, however, is not standardized because of large variations resulting from various operational factors (e.g., volume of liquid nitrogen), foam boxes (e.g., size, wall thickness), and configurations (e.g., buoyancy, and sample positioning). In most published studies, there is not reporting of these critical factors that can affect vapor-phase temperatures (Santos et al., 2013). The lack of standardized control and reporting of these variables makes comparisons problematic, and often impairs or prevents reproducibility among studies and protocols, complicating the development of germplasm repositories. Overall, this has been the state in cryopreservation for aquatic species for more than 50 years, with the field being in a repetitive cycle of inventing and re-inventing non-reproducible protocols without efficacious utility or the necessary hardware for translating research findings into reliable application at the repository level.

In recent years, however, the advent of open scientific hardware has fueled a new movement (Pearce, 2012; Pearce, 2017). Increased availability of fabrication approaches such as three-dimensional (3-D) printing now makes it possible for rapid prototyping of devices and open-access fabrication of hardware at an economical cost. This type of approach has been integrated into the field of sperm cryopreservation of aquatic species (Tiersch and Monroe, 2016; Hu et al., 2017). Open technologies can allow global accessibility of customized, standardized devices (Liu et al., 2020), enabling an aggregated high-throughput model for repository development.

3. Open technology concepts

3.1. Compiling existing terminology into a new framework

In this review article, the term “Open Technology” is used with some legitimate trepidation. Terms such as “open”, “open-source”, “free”, “free and open”, and “libre” are currently used widely with almost complete interchangeability across multiple spectra of divergent enterprises. Looking backward, much of the basis for the thinking behind the concept of “openness” can be traced to software developers which ironically began with sharing of source codes in the 1950s and 1960s with subsequent movement into development of proprietary systems as computing technologies were commercialized (Hippel and Krogh, 2003). In reaction, the “open source” or “free software” movement developed to address loss of access to programming code that was often communally developed (Raymond, 2001). In this case “free” refers to liberty (possession of defined freedoms), not price (i.e., gratis, without cost). As such, open-source software can become the focus of large distributed communities of developers who share their modifications and improvements, and thus can be viewed as two-way or omnidirectional communication (Corbly, 2014).

3.2. Features of open-technology systems

Biologists are already familiar with some open platforms, such as GenBank^® (ncbi.nlm.nih.gov/genbank/), where researchers can access and contribute to genetic data through community efforts (Benson et al., 2012). Overall, we use the term “open technologies” when referring to capabilities, designs, or hardware that are intentionally made widely accessible (e.g., by low cost, or minimal expertise requirements) within communities to address recognized problems. This is based on development of mechanisms for easy and self-reinforcing communication (e.g., websites, videos, shared web servers) that can provide training, enable evaluation, and support innovation. Adoption of these approaches to address germplasm repository problems can mediate transformation of the adopters into “user communities”. The problems associated with linking innovation to the actual problems in a field can be addressed in several ways including industrial engineering concepts based on “user-centered design” (Salvo, 2001). As user communities mature, members can transition into “maker communities” (Morreale et al., 2017) by developing the capabilities to fabricate their own devices, and eventually designing, customizing, or improving them with the emergence of “developer communities”. To function well, the interactions of users, makers, and developers are facilitated by application and adherence to licensing agreements, for example those put forth as indicated above through organizations such as Creative Commons.

In addition, open processes facilitate much-needed reproducibility in multiple ways. For example, open hardware can be developed and distributed to be standardizable (based on shared hardware and procedures) or harmonizable (based on similar results). These efforts become more powerful if mechanisms are put forward to share data such as freezing curves, and information such as population phenotypes or genetic markers in relation to the germplasm being frozen and stored in a repository (Fig. 1). The advent of inexpensive custom probes as subsequently described in this review paper, and routine availability of data transfer through Bluetooth^® or Wi-Fi will greatly empower these efforts.

Fig. 1.

A cryopreservation protocol becomes a pathway when integrated into a working format with high throughput and quality management capabilities including quality control (QC). Beyond this, the cryopreservation pathway (top row of activities) must be integrated with a repository (inside dashed oval) that coordinates samples, biological and genetic data, and other information to support the activities of user communities. In this example, the repository is designed to support aquaculture hatchery production of fry and fingerlings that can be distributed to farmers for growth to market size. This repository could also support genetic improvement programs including DNA marker-assisted selection. This representation can be converted to a process map, an industrial engineering approach in which the diagram provides the depiction of an interface to an underlying database that enables semi-quantitative or quantitative modeling of the processes to estimate costs, throughput, worker needs, etc. across multiple levels of scale.

Another powerful approach to standardization that results from computer engineering is to put forth “form factors” that describe the physical specifications for hardware and components (Chennareddy et al., 2017). This is important because it allows innovation to occur within standardized formats. This also allows for avoidance of the “standardization trap” resulting from communities being limited to use of a single piece of proprietary technology that comes from a single (monopolistic) source. Form factors that would be relevant to cryopreservation and repositories would include the use of specific containers such as French straws or cryo-vials, and the existing architectures for storage systems based on liquid nitrogen dewars and dry shippers (e.g., plastic goblets and canisters).

Most importantly, with use of open technologies, there must be recognition of the need to scaleup activities beyond the level of individual users or small groups (e.g., a laboratory). The concept of aggregate throughput as a means to transfer material of uniform high quality into a repository system needs to be scalable to the extent there is interaction of thousands of individuals and groups, and the findings from early efforts should not be discounted and replaced as user numbers increase. This leads to the final features of “open-source” processes, where approaches are evaluated in the context of how well the application supports the user communities. If mechanisms exist for multidirectional communication and development — the very foundation of open technologies - the system will be the networked product of many knowledgeable and motivated members.

4. Transitioning from traditional manufacturing to open technologies

4.1. Digital manufacturing

In recent years, digital manufacturing has been used by biologists in many areas, such as DNA assembly (Patrick et al., 2015), biomaterial development (Chia and Wu, 2015), labware customization (Baden et al., 2015), and study of reproduction and fertility (Ferraz et al., 2017). Manufacturing has traditionally involved engineers transforming ideas into dimensioned drawings, which were transferred to machinists with considerable skill to produce parts using processes such as tooling, joining, forming or injection molding (Groover, 2011). The advent of computer numerical control (CNC) machining in the 1940s, where the cutting tools on drills, lathes, and mills were automatically positioned, reduced the reliance on machinist expertise for precision tooling. This was followed by the rapid development of computing in the 1980s that led to computer-aided design (CAD) and computer-aided manufacturing (CAM), which bridged the gap proceeding from early design ideas to physical parts.

An example is laser engraving, where a computer-generated image or pattern is transferred to plastic or metal surfaces by ablation from a laser beam guided along an X-Y plane, resulting in precise imaging for shallow labels or even deep smooth-bore through-holes (Bogue, 2010). Originally only occurring in dedicated machine shops, consumer-level laser engravers (Fig. 2A–B) and CNC mills (Fig. 2C–D) are now available in affordable and compact formats. Desktop-sized 5-axis mills, recently fueled by crowd-sourced campaigns amongst hobbyists, can tool parts with complex curvatures such as impellers that would not be feasible to make without these CAD/CAM technologies (Stein et al., 2018).

Fig. 2.

Examples of biological applications supported by open-fabrication technologies. Laser engravers (A) (e.g., Muse³) can be used to produce patterns and cut materials, such as case components (B) with engraved acrylic sheets (transparent) for a sperm concentration measurement and adjustment device. CNC mills (C) (e.g., Sainsmart Genmitsu) can be used to produce 3-D patterns for accessories (D) made using acrylic to support electrical connections for a microfluidic system for assessment of sperm quality. Fused deposition modeling (FDM) 3-D printers (E) (e.g., Prusa Mk3) can be used to produce objects to support teaching, such as a fish skull (E, right) fabricated with PLA filament based on a 3-D scan of an actual skull (E, left). Resin-based 3-D printers (G) (e.g., Phrozen mini 4K) can be used to produce geometries for microfluidic channels and cell counting chambers (H) with feature sizes of < 100 μm.

Historically, most manufacturing processes (e.g., mills or lathes) were subtractive, where material was removed to produce the final part. In the 1980s, several precursors to 3-D printing were established, collectively known as additive manufacturing processes, where parts are created a layer at a time, including stereolithography (SLA) and fused deposition modeling (FDM) (Fig. 2E–F, described below, and in Table 1). Recognized as the earliest 3-D printing method (Hull 1984), SLA approaches (Fig. 2G–H) used light to cure and harden photosensitive resins. Light from a laser, digital light projector (DLP), or liquid crystal display (LCD) screen mask, is focused onto a specific pattern in the X-Y plane that solidifies layers of resin. The first commercial models were costly at US$300,000 (Miller 2016). Modern high-resolution printers (e.g., 8K, referring to horizontal display resolution of ~8,000 pixels) DLP or masked LCD light projection systems have replaced lasers in most SLA printers, bringing costs to as low as a few hundred US$ and accelerating the utilization by consumers.

Table 1.

A comparison of various fabrication methods relevant to open technologies comparing relatively greater-cost and lesser-cost examples within four categories. Some of these products are shown in Figure 3.

Method	Model*	Cost (US $)	XY resolution (mm)	Layer resolution	Feed material	Build volume (L/W/H mm)	Unique attribute	Community involvement****
CNC mill	Pocket NC© V2-101	6,000	0.0061	0.0061 mm	wood, plastic, soft metals	115/128/90	5-axis	Small
	Sainsmart Genmitsu 30182	300	N/A	N/A	wood, plastic, soft metals	300/180/45	No air compressor	Medium
Laser engraver/cutter	FSL© Muse3	3,500	1000	50 cm focal distance	< 6 cm	508/305/-	Touchscreen, camera	Medium
	OMTech laser©,4	500	1500	50 cm focal distance	< 3 cm	200/300/-	Interlocking wheels	Medium
Resin-based 3-D printer	Form 35	3,500	0.025	0.025–0.30 mm	FormLabs resins	145/145/185	Automatic resin fill	Medium
	Phrozen mini 4k6	350	0.035	0.01–0.30 mm	3rd party resins	132/73/129	Monochrome	Medium
FDM 3-D printer	Ultimaker© S57	6,000	0.069	0.02–0.20 mm***	Ultimaker filament	330/240/300	Dual nozzle, touchscreen	Small
	Prusa Mk3**, 8	750–1,000	N/A	0.05–0.35 mm***	ABS, PLA, and others	250/210/200	Mesh bed leveling	Large
	Creality 3D© Ender-39	224	0.1	0.1–0.4 mm***	ABS, PLA, and others	22/220/250	V-slot, allmetal extruder	Large

^*The products shown are representative models based on extensive survey for quality, service, user base, and cost.

^**Prusa Mk3 is included as a mid-range FDM printer due to the numerous options in this popular sector.

^***Layer resolution, stated by the manufacturer, with the printer using a 0.4-mm nozzle.

^****Amount of Community Involvement was assessed by survey of forums and groups in major online platforms (i.e., facebook.com and reddit.com) and is intended to estimate the number of active members on online platforms as of June, 2021: Small, < 25,000 members; Medium, 25,000–100,000; Large, > 100,000. The Creality 3D^© Ender-3 was the largest community, with ~370,000 combined members, and Ultimaker^© was the smallest with 6,000.

Website reference:

¹ pocketnc.com

² sainsmart.com

³ fslaser.com

⁴ omtechlaser.com

⁵ formlabs.com

⁶ phrozen3d.com

⁷ ultimaker.com

⁸ shop.prusa3d.com

⁹ creality3d.shop

Although developed after SLA, fused deposition modeling (FDM; also termed Filament Fabrication) printers have become the most prominent in market and are used among fields from medicine (Tack et al., 2016) to food preparation (Liu et al., 2017). These printers sequentially deposit layers of spooled filament (or pellets) material after melting in an extruder nozzle that is positioned with X-Y precision over a build plate (Fig. 2E). Hundreds of FDM-printable materials have been developed since there was initial utilization of first acrylonitrile butadiene styrene (ABS) and polylactide (PLA) thermoplastics, to now including rubber-like elastomers, electrically conductive polymers, and even ceramics (Vyavahare et al., 2020) and metals (Liu et al., 2019b). Low-cost, high-resolution stepper and servo motors control nozzle position, and have provided the impetus for utilization of FDM printers as a result of these consumer-inspired innovations. When the original FDM patents expired, open-source efforts fueled development of inexpensive printers such as the RepRap project (RepRap.org) for which there were published designs free on the internet for printers that could print many of the printer components. This diverging development path of high-end industrial printers and open-source consumer machines has rapidly advanced both the technology and adoption of 3-D printing (Horvath, 2014).

4.2. Open-license software

Open-license software movements have provided free and powerful tools for: 1) 3-D part design, 2) optimized fabrication, 3) simulation of function (i.e., modeling), and 4) application-specific programs for use with the device. These software developments are subsequently described in this review article.

Traditional barriers to creating 3-D designs have been that CAD programs developed in the 1980s (e.g., Autodesk Inc. AutoCAD, PTC Inc. Pro/ENGINEER, and Dassault Inc. SolidWorks) were costly, hardware-intensive, and difficult to master due to non-intuitive drawing interfaces. While these programs are still commonly used within engineering communities, the availability of open- and closed-source freeware with adequate capacities for conducting the desired procedures, has attracted many non-technical users.

Open-license CAD packages such as FreeCAD (freecadweb.org), RepoCAD (github.com/selftiesoftware/repocad) and Blender (blender.org) have varying capacities and difficulties incurred when utilized (Junk and Kuen, 2016). Freeware CAD programs such as TinkerCAD (tinkercad.com), SketchUp (sketchup.com), Onshape (onshape.com/en/products/free) and free for non-commercial use Fusion360 (Autodesk.com) currently have the most users due to ease of use, broad capabilities, and diversity of platforms including Mac, PC, mobile and browser versions. Along with CAD programs, the utilization of 3-D scanning techniques has vastly improved accuracy, and technologically superior scanners have entered the consumer market. By acquiring a 3-D scan of the desired object, files can easily be imported into most CAD and other software before editing and printing (Fig. 2F).

While the earliest CAD programs saved files in proprietary formats that in many cases were not compatible even with earlier versions of the same package, all of the previously described CAD programs can be used to export design files to universal standards, increasing interoperability across platforms and promoting collaboration. The most common open-access file format is “STL” (i.e., stereolithography), which contains necessary information for an additive or subtractive manufacturing device to produce objects (Hiller and Lipson, 2009). However, a new open-source file format standard named “3MF” (3D Manufacturing Format, https://3mf.io/specification/), has been developed by the 3MF consortium for specific application in additive manufacturing that includes information on materials, colors, and support structures not included in STL files. Prior to use on a 3-D printer, the design file must be processed through a “slicer program” with an algorithm to convert the design into stacked layers for printing based on specific attributes (e.g., layer thickness, extrusion speed), and parameters such as printing orientation and infill geometry, as well as the type of printer. Currently there is the greatest use for common printing applications of free open-license slicer programs such as Slic3r (slic3r.org) and Cura (ultimaker.com). Preparing files for SLA printers calls for different algorithms, also allowed for utilization of superior open-license programs, such as CHITUBOX (chitubox.com) and Lychee (mango3d.io), which offer SLA-specific functions such as the ability to change layer thickness and exposure time directly in the slicer.

Simulation techniques are used to improve the design and manufacturing process, where a part can be stressed and strained in silico (virtually) to see if it can withstand typical operational forces before it is fabricated and physically tested. Previously, standalone packages, such as Ansys Fluent (ansys.com) or COMSOL (comsol.com), that were utilized through supercomputers were the only options for such modeling. Now, many simulation programs are available as “add-ons” within CAD packages, or even in cloud computing form, where problems are uploaded to powerful servers. The graphical user interface (GUI) native to the CAD packages offers visualization of results such as where stresses occur within a part that might result in failure of device functionality. Another tool, entry-level Computational Fluid Dynamics (CFD) (e.g., Autodesk CFD) have been useful in modeling geometries of micromixers in microfluidic devices for zebrafish sperm activation (Scherr et al., 2012; Scherr et al., 2013), where there was simulation of accelerated evaluation of prototypes (Park et al., 2012; Beckham et al., 2018). While simulation techniques are not as common in free and open-license formats, several that provide for native GUIs are emerging with potential, including SimScale (simscale.com) for finite element stress analysis and OpenFoam (openfoam.com) for CFD (Marcantoni et al., 2012).

4.3. Open-access electronics

One of the key components for utilization of open technologies is the electronics that control hardware, collect sensing signals, or record data when a system requires such functionalities (Fig. 3). This has traditionally been a barrier for the scientific community not having background in electrical and computer engineering. The Arduino (arduino.cc), an open-source board developed at the Interaction Design Institute Ivrea (IDII) in Italy (Kushner, 2011; Severance, 2014), initially was released to maker communities in 2005. Arduino boards have been increasingly used in research of reproduction, such as in egg incubation systems (Kutsira et al., 2019), and mechanisms for assessment of oocyte quality (Yanez et al., 2016), and sperm characteristics (Rivas Arzaluz et al., 2021).

Fig. 3.

An example of an open electronics system consisting of an Arduino board (with a microcontroller), a motor driver, actuator modules, sensor modules, a sensor signal converter or an analog-to-digital converter, a touch panel display, a power supply unit or a battery, and wireless data transmission capability. These are “off-the-shelf” components that are inexpensive and are well known to existing user-maker-developer communities.

An Arduino board has a microprocessor with flash memory, analog and digital input/output (I/O) connections, communication interfaces, a crystal oscillator or resonator, a voltage regulator, a power jack, and connecting pins, forming a complete microcontroller system that is expandable for a broad variety of purposes. These powerful controllers were designed to be inexpensive (< US$20) and are widely available (D’Ausilio, 2012). Arduino products are maintained as open-source hardware and software under the GNU General Public License (GPL, gnu.org). Furthermore, these boards were designed to use C languages to program the microprocessors, which resulted in a marked reduction in the barrier accessing the Arduino integrated development environment (IDE) that allows for ease of programming. Variants of Arduino boards are currently available having different form factors, computing abilities capabilities, connectivity, and power requirements.

Another extremely useful open-source board is the Raspberry Pi (raspberrypi.org) distributed to support development of electronic and computing resources worldwide with a special emphasis on childhood education. Raspberry Pi is a computer without peripherals that can be utilized for an operating system, and has more computing power than Arduino. The Arduino has an on-board analog-to-digital converter and other embedded features (such as pulse-width modulation) which are not components for standard Raspberry Pi configurations, so users can choose which board best fits the user needs. Both of these inexpensive platforms are routinely updated and improved, and have large user communities.

Because Arduino and Raspberry Pi boards are widely available and used, there are many commercially available peripheral components compatible for use in a variety of applications. Adafruit Industries (adafruit.com) and SparkFun Electronics (sparkfun.com) are among retailers that sell such components and modules including sensors, robotics kits, motor drivers, communication modules, display types, and other accessories. Also, there are numerous sensor modules (Martín-Garín et al., 2018) available for measuring temperature, humidity, acceleration, pressure, force, infrared, distance, motion, and biometric signals.

Typical modern electronic circuitry is built on a printed circuit board (PCB), with components mounted or soldered to provide specific functionality. If a system requires functionality beyond available combinations, custom electronic circuits can be developed on a PCB by an interdisciplinary team having expertise in electronics (Shamkhalichenar et al., 2020). In addition, because there is a well-established PCB industry and service providers, two-dimensional electrodes and metal features can be designed and fabricated at a low cost using open-license or free PCB design software such as Autodesk Eagle (autodesk.com), KiCAD (kicad.org), LibrePCB (librepcb.org) to produce the Gerber files that define PCB manufacture.

4.4. Open data transfer and management

When an open-hardware system is used to record and collect data, transmission to a central computing system is necessary to process or store these data on local computers or cloud servers (Fig. 3). If an open hardware system is physically wired to a computer, transmission will likely occur through the universal serial bus (USB) protocol (He et al., 2014) that is widely used in consumer electronics, computer peripherals, and Arduino boards. If wireless connectivity is required for data transmission or system control, two commonly available options can be considered, Wi-Fi or Bluetooth^®. Both are standardized wireless data communication protocols and are available and compatible with most computers and smart mobile devices by default, and commercial modules are also readily available for integration with open hardware (Van den Bossche et al., 2016).

4.5. Workflow progression

There is a lack of guidance outside of traditional engineering for reporting on systematic testing and evaluations of production feasibility. For open-technology work focused on aquatic species repository development, there has been application of two broad testing phases (alpha and beta, such as those of software development) during the progression from “real-world” problems to user-ready solutions (Childress et al., 2021). For alpha (developer) testing, there has been identification of four specific phases that correspond to open fabrication concepts based on use of consumer-level technologies. For a solution to be “user ready”, beta testing by experienced (closed beta) and novice (open beta) users is performed to evaluate the overall functionality, ergonomics, and user experience to optimize hardware before mass release. As prototypes are evaluated for operation utility and performance, these prototypes are redesigned for making changes to improve function, ease of fabrication, or durability.

5. Open hardware for sperm cryopreservation

5.1. Reproducible cryopreservation

It is recognized production and quality management (QM) are two major pathways necessary for repository development that can be facilitated through utilization of open hardware (Fig. 4). First, although cryopreservation protocols have been established through basic research endeavors, laboratories often lack affordable and reliable hardware to process relevant numbers of samples for utilization of a standardized production pathway (Tiersch, 2011). Second, there are no effective systems available to enact quality management, including accurate assessment approaches, quality assurance mechanisms for prevention of defects, and quality control for elimination of inferior products (Torres et al., 2017). It is recognized that a QM pathway (Torres et al., 2016) operates separately but parallel to the production pathway with quality control (QC) and quality assurance (QA) components. Examples of recent development of open hardware for these two pathways are subsequently described in this review article.

Fig. 4.

A generalized framework for production of open hardware to support development of germplasm repositories. For cryopreservation, the production pathway (upper) includes all activities of sample processing from gamete collection through fertilization with three major phases: fresh sample handing, cryopreservation core processing, and post-freezing management. The quality management (QM) pathway (lower) includes repeated quality evaluation (QE) checkpoints that enable establishment of product quality control (QC), process quality assurance (QA), monitoring, and data transfer and management systems. We have identified 14 hardware categories for the production pathway, and six categories for the QM pathway. Abbreviations for production pathway: PSED, Precise Sperm Extraction Device; CFHD, Customized Fish Folding Device; CMAS, Concentration Measurement and Adjustment System; SFSD, Straw Filling and Sealing Device; LNPCD, Liquid Nitrogen Positional Cooling Device; SDPCD, Shipping Dewar Positional Cooling Device; DICD, Dry Ice Cooling Device; ULTFCD, Ultra-Low Temperature Freezer (−80 °C) Cooling Device; HTVD, High-Throughput Vitrification Device; CCCD, Controlled Cooling Conveyer Device; VSID, Vitrification Storage and Identification Device; SSAD, Straw Sorting Assist Device; PTD, Precise Thawing Device; STFD, Streamlined Thawing and Fertilization Device. Abbreviations for Quality Management: CDC, Concentration Determination Chip; ASC, Activation and Sensing Chip; CDS, Concentration Determination Scope; PPFMS, Pre-freezing and Post-freezing Monitoring System; CPMS, Core Processing Monitoring system; SPDM, Smart Platforms for Data Management.

5.2. Sample production pathways

The initial exploration of application of 3-D printing in cryobiology (Tiersch and Monroe, 2016) indicated there was material integrity of objects fabricated with polylactic acid (PLA) filament when immersed in liquid nitrogen. Objects 3-D printed with PLA can undergo considerable elongation prior to breaking (similar to room temperature) at liquid nitrogen temperatures (Tiersch and Monroe, 2016), indicating great potential for 3-D printing of devices and tools used in cryogenic compounds. Since this report was published, several 3-D printed devices for cryopreservation procedures have been reported, evolving in complexity from single-piece tools to complicated systems with >100 components. We have identified 14 categories of open hardware for applications in the production pathway (Fig. 4) from sample collection to fertilization. Many of these technologies are currently in development and several of those that have previously been reported are subsequently described in this review article.

Most sperm vitrification studies have been constrained to adoption of tools that were initially designed for applications other than sperm (and aquatic species). Although these tools can be used for vitrification purposes, these tools can be economically expensive (e.g., the Cryotop^® costs more than US$20/device for 2-μl samples), and are difficult to redesign, customize, securely label, and efficiently store. An early development of 3-D printed vitrification technologies for tool production (Tiersch et al., 2019) featured a single-piece loop fabricated with PLA or acrylonitrile butadiene styrene (ABS) that could support sperm samples in thin films (Fig. 6A) (Tiersch and Tiersch, 2017). More complicated vitrification devices were developed that featured elongated loops, labeling mechanisms, handles to facilitate processing, and storage by integration with existing cryopreservation vials (Fig. 5B) (Liu et al., 2021b) and straws (Fig. 5C) (Tiersch et al., 2020). This demonstrated the feasibility of standardized low-cost (<US$0.05 material cost) 3-D printed devices with practical functions including vitrification, volume control, labeling, protection, and storage within conventional systems.

Fig. 6.

Examples of open hardware for the quality management pathway. A Microfabricated Activation and Motility Chamber (MAMC) (A) with microfluidic channels for activation and assessment of fish sperm. A Microfabricated Enumeration Grid Chambers (MEGC) (B) for estimation of sperm concentration. A 3-D printed probe (C) for monitoring of cryopreservation within a 0.5-ml French straw. Probes can also be produced using customized printed circuit boards (PCB) directly for monitoring of cryopreservation phenomena.

Fig. 5.

Examples of open hardware for the cryopreservation production pathway. A vitrification quality evaluation pedestal (A) was 3-D printed to standardize visual evaluation. Vitrification Storage and Identification Devices (VSID) provide sample identification and protection by integration with existing storage systems such as cryopreservation vials (B) and straws (C). The Shipping Dewar Positional Cooling Device (SDPCD) can load 22 French straws (D) and provide a wide range of cooling rates by vertical positioning (E) within a standard shipping dewar. The Positional Cooling Platform Device (PCPD) (F) can provide standard cooling rates for freezing of batched samples. The Controlled Cooling Conveyor Device (CCCD) (G) can transport samples during cooling and automatically drop them in liquid nitrogen after reaching target temperatures.

In addition, 3-D printed devices were also developed for conventional slow cooling (i.e., equilibrium freezing). A Shipping Dewar Positional Cooling Device (SDPCD) (Childress et al., 2021) was developed for freezing and storing samples on-site (Fig. 6E). The SDPCD comprised eight 3-D printed components and a commercially available metal spring (total US$8 material cost) with various standardized cooling rates ranging from 1 to 68 °C/min. A major advantage with the design of the SDPCD was the radial arrangement of straws, preventing contact with each other or the inner wall of the dewar. In addition, a quick-release mechanism enabled direct release of straws after freezing, eliminating the risk of sample damage due to handling of semen samples outside of dewars. Ejecting straws directly into the dewar canister (or a large goblet) allowed for samples to be frozen in the field and easily transported in the same container to a central facility for sorting and long-term storage.

In addition to liquid nitrogen shipping dewars, polystyrene foam (Styrofoam) boxes have been used to contain liquid nitrogen to cool samples with nitrogen vapor (Adams et al., 2004; Irawan et al., 2010). A 3-D printed Positional Cooling Platform Device (PCPD) (Fig. 6F) was developed (Hu et al., 2017) to produce reproducible cooling rates (4 to 40 °C/min). The PCPD included 14 3-D printed components, two pieces of Styrofoam to provide floatation, and nested Styrofoam boxes. The PCPD could accommodate various containers (i.e., 0.25 and 0.5-ml French straws, 0.5 and 2-ml vials) with a material cost of US$5.

An instrumented Controlled Cooling Conveyor Device (CCCD) (Liu et al., 2021a) (Fig. 5G) with electrical control, motorization, and a large number (> 100) of individual 3-D printed components was developed for continuous cryopreservation of non-batched samples (Liu et al., 2021a). There were no significant differences in post-thaw motility of sperm from ornamental (koi) common carp (Cyprinus carpio) among samples frozen with the CCCD and those frozen with a commercial programmable freezer. This is indicative of the potential to develop devices with advanced and complicated functionalities made accessible with integrated open-access electronics, such as sensing, monitoring, automated control, and compatibility with the Internet of Things.

5.3. Quality management pathways

Integration of a QM pathway along with the sample production pathway is indispensable for establishment of practical repository programs. Innovative hardware was developed for standardized quality evaluation for QC purposes and process monitoring for QA purposes. Measurement of sperm motility and concentration are two of the most important quality assessment variables (Torres and Tiersch, 2018). The common practice of pre-set dilution (normalization) of sperm samples for freezing (Draper and Moens, 2009) without estimating or adjusting for sperm concentration can compromise reproducibility because cell concentration affects the efficacy of various cryopreservation procedures in a variety of ways (Dong et al., 2007; Draper and Moens, 2009). Motility assessment lacks standardization among groups, and can vary within laboratories because manual methods are subject to human-induced variation. Computer-assisted sperm analysis (CASA) has improved reproducibility as compared with when there was manual estimation (Verstegen et al., 2002), but even with CASA, due to variation in technique, there is a lack of standardization resulting in inconsistent motility analysis. Small sample volumes (1–2 μl), and a short motility duration (e.g., < 1 min) (Tiersch, 2011) add to the complexity of these difficulties (Alavi and Cosson, 2006; Hagedorn et al., 2009). We have identified six categories of open hardware for utilization in the QM pathway from sample collection to fertilization (Fig.4). Published examples are subsequently described in this review article.

Sperm from many aquatic species require activation by mixing with hypotonic or hypertonic media and remain actively motile only for seconds. As such, development of rapid mixing configurations (Park et al., 2012; Scherr et al., 2012) is essential for success of microfluidic activation devices. A Microfabricated Activation and Motility Chamber (MAMC) micromixer (Fig. 6A) was molded in polydimethylsiloxane (PDMS) bonded to a glass substrate, allowing for rapid analysis (Beckham et al., 2018). This demonstrated feasibility for development of microfluidic systems for quality assessment of sperm from aquatic species and signals a pivotal advancement in streamlining methods for standard, consistent, and rapid assessment for reproducibility of research and repository development.

To assist routine concentration assessment at an affordable cost, Microfabricated Enumeration Grid Chambers (MEGC) were developed using microfabrication techniques and zebrafish sperm for testing (Liu et al., 2020). These MEGC prototypes (Fig. 6B) were composed of a PDMS coverslip with grid patterns (100 × 100 μm) and a PDMS base platform to create a known volume with a 10-μm height to restrict sperm cells to a single layer. Counts estimated with utilization of these prototypes were comparable to a those determined using a commercially available Makler^® chamber (Christensen et al., 2005). The material cost for a MEGC was < US$0.1 compared to ~US$100 for a standard hemocytometer, and ~US$700 for a Makler^® counting chamber.

Another example is customizable sensors based on 3-D printing techniques can be developed for monitoring of cryopreservation phenomena (Shamkhalichenar et al., 2019). Sensing probes (Fig. 6C) (<US$20 material cost) were designed to fit within standard 0.5-ml French straws. Phase-transition phenomena were detected by analyzing electrical resistance changes (Shamkhalichenar et al., 2019). A portable platform was developed for monitoring of freezing and thawing based on electrical impedance which included a PCB-fabricated probe (Shamkhalichenar et al., 2021) (Fig. 6D). The material cost of this system was ~US$300/unit compared to >US$3,000 for commercial systems with comparable capacities for conducting these procedures.

6. Participation in development of open technologies

6.1. Community-based standardization

Although technology development provides for a common goal, it will be difficult to develop widely usable approaches without the collaboration of people or groups with diverse expertise and abilities. Multidisciplinary cooperation is a starting point, because this results in knowledge being combined from different disciplines, but with each person remaining within their own field. Even more effective, is an interdisciplinary group, in which knowledge and methods are drawn from individuals focused on multiple disciplines, and synthesized and integrated in a new knowledge approach (Collin, 2009). If the thoughts of individuals in developer and maker communities are not well integrated with the users, the original developer may be able to make changes and provide updates until other developers emerge (Fig. 7). Subsequently, remaining versions of the hardware will decrease in number in a convergence phase that can result in opportunities for those leading community-based standardization efforts. A process such as this would draw on the innovation and capabilities of large groups to produce refined versions of open hardware. This could also be an impetus for the expansion of the pool of developers and makers participating in innovation for other types of hardware.

Fig. 7.

A depiction of how innovation can be introduced into differential application and later converge into standardization. Open hardware can be evaluated by early adopters among user-maker-developer communities, and it diverge through modification based on different applications. Convergence is needed to integrate the modifications into a standardized approach at the community level.

6.2. Open file-sharing platforms

Some of the most well-known file-sharing platforms include Github (github.com), Thingiverse (thingiverse.com), Gitlab (gitlab.com), HackADay (hackaday.com), Wikifactory (wikifactory.com) and Instructables (instructables.com). Each of these platforms has unique characteristics and intended users. Github is a web-based application that hosts projects and files, originally developed for software collaborations, that has expanded to other types of projects; accounts for open-source projects are free, and paid accounts are available for commercial projects. Thingiverse allows the sharing of digital designs, particularly 3-D printing files, but also files for other technologies. Gitlab is a repository management tool that can track code progression, and it is used mostly for software. HackADay allows communication and exchange of information among makers in a less structured way than the other platforms. Wikifactory is a collaborative product development social platform, based on software development. Instructables is general platform (Autodesk, Inc.) and includes free access to thousands of tutorials, workshops, and projects. When evaluating a platform, it is important to determine the accessibility to the target community, including price and resources needed for access, ease of use, objectives, and type of protection needed. As with all other aspects of technology sharing, the real value is not only in the hardware, but in the possibility that it has utility for end users.

6.3. Economics

The production of open hardware is not simply altruism or “giving things away” as is often perceived. It is in fact an emerging complement to traditional “bricks and mortar” business models and conventional intellectual property mechanisms such as patenting and licensing. New business models are arising from open-source concepts including kit suppliers, specialty component suppliers, and calibration and validation services (Pearce, 2017). Economic tools to quantify the value of open hardware (Pearce, 2015) including production of classroom learning aids (Gallup and Pearce, 2020) are being developed and indicative of the feasibility that open fabrication of even relatively simple devices can be utilized for accruing millions of dollars of economic value representing orders of magnitude in larger economic gains in comparison to when there is utilization of traditional proprietary mechanisms (Pearce, 2016). Because this is an emerging area, entities such as granting agencies and leadership for scientific journals are only beginning to recognize the potential in these regards (Gibney, 2016).

7. Conclusions

As we prepared to participate in this Festschrift issue honoring Dr. Garner, the thoughts of one of us (TRT) went back through 35 years of flow cytometry launched with a Coulter EPICS V, a sprawling multi-room “beast” of an instrument. That machine required constant “jack-of-all-trades” maintenance and came from a time when tinkerers, engineers, and biologists regularly shared their skills to advance the technology. The refined instruments and reagents of today allow us to remain securely ‘cocooned’ within our specialties, which makes it important for us to reflect on the career of Duane Garner as a way of looking forward by remembering the past. We hoped to capture that spirit by honoring Garner with an article that calls attention to the power of interdisciplinary technology development as an approach to address serious problems in sperm cryopreservation that limit our progress in developing germplasm repositories for aquatic species.

Open interdisciplinary cooperation has fueled previous advances that have revolutionized science and technology such as personal computing, the internet, social media, digital photography, and of course open-source programming. Open hardware can offer benefits to every program of study highlighted in the many articles of this special Festschrift issue. We hope that this article can help raise awareness, stimulate future collaborations, and spread the concept of open technology approaches to new user-maker-developer communities to support reproductive biology, cryopreservation, and the protection of genetic resources of aquatic species.

Highlights.

• Cryopreservation research and protocols themselves do not translate to repositories.

• Emerging open technologies can support reproducibility in repository development.

• Standardization with open hardware can lead to community-level standardization.

• We recognize 20 categories of open hardware for processing and quality management.

• Open hardware offers economic incentives for practical repository development.