bioRxiv preprint doi: https://doi.org/10.1101/796516; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. Chi.Bio: An open-source automated experimental platform for biological science research Harrison Steel1, Robert Habgood1, Ciaran´ Kelly2, and Antonis Papachristodoulou1 The precise characterisation and manipulation of proteins) which are fundamental to many experimental in vivo biological systems is critical to their study.1 studies. However, in many experimental frameworks this is made challenging by non-static environments during cell To address these challenges we developed Chi.Bio, growth,2, 3 as well as variability introduced by manual a parallelised all-in-one platform for automated sampling and measurement protocols.4 To address characterisation and manipulation of biological systems. It these challenges we present Chi.Bio, a parallelised is open-source and can be built from printed circuit boards open-source platform that offers a new experimental (PCBs) and off-the-shelf components for ∼ $300 per device. paradigm in which all measurement and control The platform comprises three primary components (Fig. actions can be applied to a bulk culture in situ. In 1a); a control computer, main reactor, and pump board. The addition to continuous-culturing capabilities (turbidostat control computer can interface with up to eight reactor/pump functionality, heating, stirring) it incorporates tunable pairs in parallel, allowing independent experiments to be light outputs of varying wavelengths and spectrometry. run on each. It also hosts the platform’s operating system, We demonstrate its application to studies of cell growth which provides an easy-to-use web interface for real-time and biofilm formation, automated in silico control of control and monitoring of ongoing experiments. The main optogenetic systems, and readout of multiple orthogonal reactor contains most of the platform’s measurement and fluorescent proteins. By combining capabilities from actuation sub-systems (Fig. 1b), which operate on standard many laboratory tools into a single low-cost platform, 30 mL flat-bottom screw-top laboratory test tubes (with Chi.Bio facilitates novel studies in synthetic, systems, a 12 to 25 mL working volume). All measurement and and evolutionary biology, and broadens access to actuation systems (with the exception of the heat plate) cutting-edge research capabilities. are non-contact, minimising sterilisation challenges and allowing test tubes to be hot swapped during operation. Biology faces a crisis of reproducibility, caused in part by Each reactor can accept up to four liquid in/outflow tubes, a lack of control over conditions experienced by cells prior which are driven by peristaltic pumps housed in the reactor’s to and during experiments.5 Biological systems are often dedicated pump board. The platform as a whole is entirely characterised using batch-culture methods,6, 7 which make modular (the three components inter-connect via micro-USB isolating a cellular sub-system’s behaviour from that of its cables), allowing it to be tailored to a wide variety of host challenging.3, 8 To improve the robustness of biological experimental configurations. data an ideal experimental setup would provide a controlled, static environment in which culture parameters such as Experimental techniques which exploit interactions nutrient availability and temperature are regulated,9 and between light and life, such as light-sensitive proteins, perform frequent and accurate measurements in situ. This optogenetics, and fluorescent reporters, are ubiquitous in can be partially achieved using continuous culture devices biological research.16, 17 Chi.Bio contains an array of optical such as a Turbidostat, which dilutes cells during growth to outputs and sensors to support these techniques (Fig. 2a). maintain a constant optical density (OD). In recent years a A 650nm laser is used for optical density measurement number of Turbidostat platforms have been developed, and (calibrated against a Spectrophotometer, Note S5), and are beginning to find widespread applications in systems, is driven by an analogue feedback circuit (Fig. S6) to synthetic, and evolutionary biology.10–15 However, in many provide stable, temperature-insensitive readings (Note cases these devices are not easy to build/obtain, require S6). Optogenetic actuation and excitation of fluorescent interfacing with external hardware (such as incubators), are proteins employs a focused high-power 7-Colour LED inflexible for applications beyond their designed purpose, with six emission bands across the visible range and a and lack in situ measurement and actuation capabilities (such 6500K white output (Fig. 2b,c), and a separate 280nm UV as optogenetic actuation or measurement of fluorescent LED is included to stress/sterilise cells. Each LED has an independent current-limiting pulse width modulation (PWM) 1Department of Engineering Science, University of Oxford, Oxford, OX1 driver (Fig. S6), allowing its intensity to be regulated over 3PJ, UK. 2Department of Applied Sciences, Faculty of Health and Life three orders of magnitude (Fig. S7), and is thermally coupled Sciences, Northumbria University, Newcastle upon Tyne, NE1 8ST, UK. e- mail: (fharrison.steel,[email protected]). This research was supported to the outside of the device to reject heat when operating at by EPSRC project EP/M002454/1. high intensity. The combined LED implementation provides bioRxiv preprint doi: https://doi.org/10.1101/796516; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. a 1cm b PC Start Stop Pump 1 Pump 2 Pump 3 Pump 4 Data UV LED Spectrometer Time 7-Colour LED Infrared Thermometer 650nm Laser Fresh Waste Input 1 Input 2 Media Ambient Thermometer Microcontroller Magnetic Stirring Fig. 1: (a) The Chi.Bio platform comprises a control computer (left), main reactor (centre), and peristaltic pump board (right). The system is open-source and can be constructed for ∼ $300 using only PCBs and off-the-shelf components. Scale bar indicates 1 cm, giving the main reactor dimensions of 11.5×5.3×5.3 cm. (b) Schematic of sub-systems and interconnections. A lab computer or network connects to the control computer, which runs the platform’s operating system and can interface with up to eight reactor/pump pairs in parallel. Each reactor has a 12 to 25 mL working volume and contains a range of measurement and actuation tools for precise in situ manipulation of biological systems. These include a UV LED, a 650nm laser (for OD measurement), seven coloured LEDs in the visible range (for optogenetics and fluorescence measurement), and a spectrometer. An infrared thermometer and heat plate are used to regulate temperature, and the culture is agitated using magnetic stirring. Each reactor has a modular pump board with four direction- and speed-controllable peristaltic pumps. For a detailed descriptions of each hardware sub-system see Notes S1-S4. optical outputs that exhibit minimal power and spectral and spatial variations in culture density. variability between devices or environmental conditions (Note S7). Culture temperature is measured non-invasively by a medical-grade infrared thermometer, which is accurate to Measurements of light intensity are performed within ± 0.2 ◦C for temperatures near 37 ◦C. There are also Air the device by a chip-based spectrometer with eight optical temperature thermometers (± 0.5 ◦C accuracy) within the filters covering the visual range, as well as an un-filtered main reactor and on the control computer for monitoring the “Clear” sensor (Fig. 2d). Multiple wavelength bands can be surrounding environment. Temperature change is actuated measured simultaneously, each with electronically adjustable by a PCB-based heat plate, capable of heating a 20 mL gain and integration time. The spectrometer is set up to culture at up to 2.0 ◦C min−1. Below the heat plate is a perform temperature and long-term baseline calibration magnetic stirring assembly built upon an off-the-shelf fan, using a dark photodiode prior to every measurement. which has an adjustable stirring rate (Fig. 2e) and can be Typical spectrometer measurements (e.g. of fluorescence) used with standard laboratory stir bars. The main reactor are reported as the ratio of light intensity measured at the also includes an external expansion port, which provides fluorescent protein’s emission band to the total intensity of power and a digital interface for user-built add-ons to the excitation source; this ratiometric measurement mitigates Chi.Bio. the impact of differing excitation intensity between devices bioRxiv preprint doi: https://doi.org/10.1101/796516; this version posted October 7, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. c a 395 457 500 523 595 623 b 1 Laser Laser 395 457 500 0.5 6500K UV LED Emission Intensity 523 595 623 6500K 0 OD Laser 350 400 450 500 550 600 650 700 Optical Outputs Wavelength (nm) 7-Colour LED d 410 440 470 510 550 583 620 670 e 1 Spectrometer Heat Plate Infrared 0.5 Clear Thermometer Magnetic Stirring 0 Measurement Sensitivity 350 400 450 500 550 600 650 700 0.3 0.6 1.0 Wavelength (nm) Stir Rate f 8x Digital g Time (s) 0 30 60 Multiplexing Real-time Data Stir Off On Reactors Measure OD FP 1 ... FP n Temp PC User inputs Control + Pumps Computer Pump Calculate Out In [ [ [ Actuate Optogenetic / Chemical / UV actuation User Interface Hardware OS Digital Bus [(HTML, Javascript) [ (Python) [ (I²C) Data Write Data Update UI 0.4 0.4 Measured 40 h Measured i Measured j 25 ° C 37 ° C Set-point Set-point 0.35 38 0.35 0.3 1 1 - 36 - C) ° 0.3 0.25 34 OD cm OD cm 0.2 Temp ( 32 0.25 0.15 Dither (top) 30 Dither (bottom) 0.2 0.1 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 Time (h) Time (h) Time (h) Fig.