Patent Application
20080006653
1. Field of the Invention
This invention relates to automated methods and systems for small volume liquid dispensing, and more particularly, to vision (imaging device/camera) guided and solenoid valve dispensing systems and methods usable for a variety of biomedical and other applications.
2. Description of Related Art
Nanoliter liquid handling has become an important technology area because increasingly there are requirements for handling very small volumes of a variety of liquids in high throughput drug discovery, medical diagnostics, basic scientific research, and many other biomedical applications. Historically, small volume liquid handling applications were limited to dispensing in pre-defined row column matrices embodied by 96, 384, and 1536 well microtiter plates or higher density arrays. As the plate densities increased, the desired dispense volumes decreased, and technology development has focused mainly on improving small volume liquid handling over the range of 5-100 nanoliters.
Two basic technologies for small volume liquid handling are contact printing and non-contact liquid handling. Both technologies may employ automated robotic systems. Contact printing can work well with small volumes of liquid, but because of physical contact can damage the surface of a substrate or can destroy an object on the surface of a substrate, such as tissue. Non-contact liquid handling systems for very small and small volumes (i.e., picoliter to nanoliter) typically incorporate piezo actuators similar to inkjet techniques, or very high speed solenoid valves, depending on the desired dispense range. Theoretically, piezo based liquid handling techniques should be competitive with contact printers in microarray dispensing; however, piezo dispensers have issues related to clogging, cross contamination, sensitivity to barometric pressure, and other practical issues, especially incompatibility with certain chemical and biological materials.
Recently, automated small volume liquid handling has become important because of the need for high speed in large scale, high throughput operations. Automated small volume liquid handlers repeatedly dispense a reagent, solvent, or sample into an object container (e.g., high density microtiter plate, or array substrate). Existing liquid handlers are limited by reliability and ease-of-use issues which negatively impact both performance and function.
Although the trend for miniaturization for liquid handling has increased the requirement for precision of automated robotic systems, emerging applications in the biomedical field also require increasing flexibility in placement and patterning of drop depositions (e.g., tissue mass spectrometric imaging applications). In addition, two recent trends are driving technology development for automated small volume liquid handling: (1) There is increasing need to dispense low or sub-nanoliter volumes to conserve rare or precious materials; and (2) there is increasing need to dispense small volumes at specific or operator defined locations for a variety of regular and irregular substrates (e.g., tissue microarrays or tissue slices dried onto matrix-assisted laser desorption/ionization (MALDI) targets). The performance and reliability of a solenoid valve based dispenser has heretofore generally been limited to reliable dispensing over the range of 10's to 100's of nanoliters.
For the foregoing reasons, there is a need for development of reliable, easy-to-use, accurate, and flexible small volume liquid handling techniques and devices. Among other areas, a variety of biomedical applications may benefit from such techniques and devices, especially in the fields of genomics, proteomics, and high content cell biology studies.
The present invention relates to vision-enabled, small volume liquid delivery systems and methods of using same for automation of complex laboratory workflows and especially useful for automating front-end sample preparation for matrix assisted laser desorption/ionization (MALDI) mass spectrometry analysis.
In one aspect, the present invention relates to a system that integrates vision with small-volume dispensing comprising:
Another aspect of the present invention relates to an imaging device guided small volume liquid handling apparatus for use in deposition of liquids onto an object or substrate in accordance with a specified pattern comprising:
In yet another aspect, the present invention relates to a method for dispensing small volumes of liquid comprising:
A still further aspect relates to a camera calibration and image processing method comprising:
In another aspect the present invention relates a method for calibrating droplets, the method comprising:
In yet another aspect, the present invention relates to a dispensing unit comprising:
A method for dispensing small volume liquids with imaging thereof, the method comprising:
Various other aspects, features and embodiments of the invention will be more fully apparent from the ensuing disclosure and appended claims.
The present invention will be described in terms of specific, example embodiments. It is to be understood that the invention is not limited to the example embodiments disclosed. It should also be understood that not every feature of the devices or methods described are necessary to implement the invention as claimed in any particular one of the appended claims. Various elements, steps, processes, and features of various embodiments of devices and processes are described in order to fully enable the invention. Throughout this disclosure, where a process or method is shown or described, the steps of the method may be performed in any order or simultaneously, unless it is clear from the context that one step depends on another being performed first.
In the Figures herein, unique features receive unique reference numerals, while features that are the same in more than one drawing receive the same reference numerals throughout. Further, certain terms of orientation may be used, such as “front,” “back,” “interior,” “top,” “bottom,”“left,” “right,” “vertical,” and “horizontal.” These terms are generally for convenience of reference, and should be so understood unless a particular embodiment requires otherwise.
Referring now to the drawings, an embodiment of an imaging device guided automated liquid handling system 100 in accordance with the present invention is shown in
The scope of the invention is not intended to be limited by materials or specific manufactured parts listed herein, but may be carried out using any materials that allow construction and operation. Materials, parts, and dimensions depend on the particular application. In general the materials of the components may be as appropriate for pneumatic, hydraulic, and electronic parts, as selected by one of ordinary skill in the art. All dimensions discussed herein are by way of example.
For the operation, there are several steps, including camera calibration, implementation of the dispensing protocol, nozzle calibration, image processing and liquid handling spot generation, liquid handling, and post-liquid handling.
As shown in
As shown in flow diagram of
All the images acquired will be reconstructed afterwards via warping the raw image from the target center of gravity array to the robot-moving array. Warping is a process that deforms images and in which positions in one plane are mapped to positions in another plane. The result may include removal of optical distortions that can result from camera lenses or viewing perspective. Images may be registered within an array. Multiple images may be aligned.
The center of gravity of the target may be determined from a distorted image. The center of gravity from the image of the target may be calibrated first in the image and then the center of gravity of the target may be calculated from the initial calibrated image data. A more accurate position calibration may be achieved than with conventional methods. For example, in one embodiment where the target is a circle about 1/10 the diagonal of the FOV, the inaccuracy of the center of gravity position from the image distortion of the target is about 1/10 of the target position inaccuracy in the whole image. Thus, about 90% of the geometric distortion will be removed after the first iteration. If higher positional accuracy is required, a second iteration of camera calibration can be performed with another 90% distortion removed from the image. Also, multiple iterations of camera calibration will further improve the imaging geometric accuracy.
Thus, with respect to image geometric accuracy, the imaging device/camera calibration process and imaging processing of the present invention may reduce geometric distortion of the image produced by the imaging optical system. On the other hand, many conventional machine vision systems find the feature position from the raw, distorted image captured by the camera. Then a more accurate position is calculated from a warp function with warp coefficients obtained from a process similar to that described above. However, since the calculation of the center of gravity is in raw data, there will be associated distortion and may be lower accuracy of the calibrated center of gravity position as compared with the method of the present invention.
Image processing is performed with the use of software from the Intel® Integrated Performance Primatives, Image Processing Function Domain (INTEL is a registered trademark of Intel Corporation), as may be selected and used by one of ordinary skill in the art.
An initial washing cleans residual material from a previous use and detects clogging in the liquid handling flow path. It may be necessary to wash the outer nozzle 316 using an ultrasound cleaner. When desired, the robot 101 may automatically move to an embedded ultrasonic washer 130, as shown in
One example of a prime fluid 318 is distilled water. Dispense fluid 320, on the other hand, may be harmful to the dispense valve 302. Accordingly, during operation the dispensing fluid 320 is stored in the reservoir tubing 404 and is not allowed to enter the dispense valve 302 in order to prevent clogging and physically or chemically interacting with the internal components of the valve 302. In tissue mass spectroscopic applications, Sinapinic acid (3,5-dimethoxy-4-hydroxycinnamic acid, or SA) is typically used for high molecular weight proteins while -cyano-4-hydroxycinnamic acid (HCCA) is more commonly employed for low molecular weight proteins or peptides. In operation of one embodiment, 10-30 mg/mL of SA concentrations can be used.
In one embodiment a 15 mm long, 0.5 mm I.D. Teflon® (polytetrafluoroethylene) tube may be used to connect the nozzle 316 and the valve orifice and serve as the dispense fluid reservoir (TEFLON is a registered trademark of E.I. du Pont de Nemours and Company). About a 2.5 microliter volume in total may be formed based on use of this tube underneath the dispense valve orifice.
Reservoir tube 404 size may be used to control volume, along with the time the dispense valve 302 stays open or the time the pump or syringe 314 stays on, which is controlled by software. For lower volume nano liquid handling applications, to avoid the mixing interface, only 1 to 2 microliters of fluid may required for dispensing, which is configurable by the operator. For 1 nanoliter droplet size, about 1,000 to 2,000 spots can be dispensed. Longer Teflon tubing 404 can be used if larger droplets need to be dispensed. In one embodiment, after aspiration the system's pneumatic and hydraulic system's gas and liquid border will remain located just above the dispense valve, and may be located at a branch of the Y connector. The gas may be air, as generally referred to herein, or may be another gas as selected by one of ordinary skill in the art; reference to air, air pressure, and air pressure valves should be understood to refer to any gas. Thus, a certain small volume of prime liquid will stay above and below 806 the dispense valve 302 during dispensing (i.e., minimally fluid filled), which optimizes the valve response frequency and consistency of the dispense performance. Again the volume of dispense fluid is controlled 808 by software, and at the appropriate time aspiration is ended. The motive force of dispensing is provided by the pressurized minimally fluid filled system upstream of the dispense valve 302. Subsequently the air pressure valve 312 opens. The dispense valve 302 is backed by air pressure, the level of which is controlled by the software.
Prior to dispensing, a pre-dispense process beneficially may be applied to enhance liquid dispensing reliability. In the pre-dispense process, a relatively large pressure may be applied and the dispense valve 302 opened to initially dispense 812 a relative large droplet. At the same time, the dispense valve control pulse will be wider so that the dispense valve open time will be longer. The larger droplet dispense may contribute to reliable dispense liquid flow operation. After a few larger droplets are dispensed, in one embodiment in part of the slide holder but generally in any other location than the tissue, the valve control pulse width and dispensing pressure are reduced gradually 814 until the desired smaller droplets are dispensed. Then pressure, for example from 0 to 6 psi, and the valve opening pulse widths are set by software to the final dispensing parameters.
As illustrated in
In the initial dispense process, discussed below, nozzle dispense positioning can be further verified and calibrated during each nano-dispensing run. Real-time nozzle calibration may contribute to accurate and reliable liquid handling positioning.
The high precision robot 101 moves the camera 114 through X,Y space and acquires multiple partial images of the object. Software then creates an object image using a tiling process in which these multiple partial object images are assembled to create the object image.
Initially, an operator mounts an object (e.g., a MALDI target containing a dried thin section tissue sample) for dispensing onto an object or target plate holder 124, as shown in
The second method, shown in
In one embodiment, illuminated object images are obtained by an imaging device 114, such as a camera or image sensor, through an optical lens which may or may not be fitted with emission or other filter devices. Image data obtained from the camera or sensor is transferred to a computer for image storage and processing. Through manual, semi-automatic, or interactive and automated protocols, a droplet deposition pattern is generated for liquid handling. An operator can define a dispense pattern for the substrate or object image, or use an imported pattern or list of coordinates for liquid handling. An operator also can use computer software to interactively place, resize, drag, drop, and rotate the pattern to orient or align with the substrate or object, such as shown in
The dispense volume may be determined by the combination of upstream system pressure, as discussed above, and the spike and hold waveform driving the valve opening and closing generated by the electronics. A voltage spike is applied to open the dispense valve 302, and is maintained briefly. Then the voltage is dropped and held to avoid burnout of the valve before applying another spike to open the valve again. An interlaced dispensing mechanism may be used whereby alternate droplets may be deposited and allowed to dry before the system deposits the intervening droplets, which may result in denser placement of droplets. In addition, the combination of vision, dispense calibration and compensation, and a high precision robot platform may allow the system to repeatedly dispense the same or different liquids multiple times in an overlay at the same location with or without drying between dispense cycles.
In order to contribute to reliable dispensing, especially for some difficult liquids (e.g. MALDI matrix), an initialization process may be provided that is software controlled and coordinates the duration of the solenoid valve 302 drive circuit electronics' spike and hold pulse with the level of system pressure in the dispensing liquid flow path. The liquid dispensing can be carried out with high speed from location to location by moving the robot's X,Y linear motors to the specified X,Y droplet deposition coordinates. After a specified number of dispenses, a washing process, discussed below, may be introduced to clean the inside and outside of the dispense nozzle. These washing steps may be helpful since partial or full clogging of the nozzle will reduce the system reliability.
The D input to the first trigger 1204 may be from an AND gate 1206 which logically ANDs the D1 input and an inverted output (
The nano dispense valve control circuit 1200 may allow the robot programmable control of the dispense valve opening pulse width to vary from about a microsecond or less to about a second or more. In the circuit, the voltage frequency converter (V/F) 1202 is used. The V/F converter 1202 produces a square wave train. In operation, when the square wave at the J or clock input to the first trigger 1204 goes from low to high and if the input D1 is high, the Q output of the first trigger 1204 will go to high. The Q output of the first DJ trigger 1204 being high and the D2 input being low will set the complimentary or inverted output (
When D2 is on or high, the output pulse DNDV width will be about the same as the pulse width of the output signal of the voltage frequency converter 1202. The DNDV pulse width may vary from about several hundred microseconds to several seconds as programmably controlled by the robot, which can be used for the aspiration process as illustrated in
The about 5-15 volts DC bias voltage may also be connected to pins 4 and 8 of the chip 1414. Pin 8 of the chip 1414 may be connected to pins 6 and 7 by a resistor R11415. Pin 6 may be connected to ground potential by a capacitor C11416 and pin 5 may be connected to ground potential by another capacitor 1418 that may be about a 0.01 micro Farad capacitor or the like. Pin 1 of the chip 1414 is also connected to ground potential.
Pin 3 of the chip 1414 may be connected to a base of a third transistor 1420 by a base resistor 1422. The third transistor 1420 may be a PNP transistor and the base resistor 1422 may have a resistance of about 10 kilo ohms. The emitter of the third transistor 1420 and the emitter of the second transistor 1404 may be connected to ground potential. The collector of the third transistor 1420 may be connected to a collector voltage V1 by a collector resistor 1424 that may be about 10 kilo ohms. The collector of the third transistor 1420 may also be connected to the base of a fourth transistor 1426 by a base resistor 1428. The fourth transistor 1426 may be a PNP transistor. The base resistor 1428 may be about a 4700 ohm resistor to provide proper bias of the fourth transistor 1426. The collector of the fourth transistor 1426 may be connected to the cathode of a diode 1429. The anode of the diode 1429 may be connected to the anode of a Zener diode 1430. The cathode of the Zener diode 1430 may be connected to the collector of the second transistor 1404. The valve control voltage or DNDV may be the voltage across the cathodes of diode 1429 and Zener diode 1430 which is applicable to the nano dispense valve 1432. The nano dispense valve 1432 is represented electrically in
As previously discussed, for liquid handling, reagents, samples, or solvents can be aspirated from a variety of source containers such as tubes, microwells, or Eppendorf type containers. After the pre-dispense process, the robot 101 will move the dispense head 110 to the object's first dispense location and immediately commence the object dispensing operation. The dispense sequence is software controlled and uses the droplet locations determined during image processing. The liquid can be dispensed at the specified X,Y droplet coordinates. The Z coordinate can be predetermined for an object type and stored in a configuration database. The dispense locations are stored in an XML file that the robot software accesses to determine the order and location for each dispense.
The computer software control may provide improved reliability for nanoliter liquid handling, while, at the same time, may limit the need for human intervention, increase reproducibility, enable walkaway automation, and decrease the number of potential human errors.
Based on the software configuration, an operator can specify the number of dispenses before washing the liquid handling nozzles 316. For example, an operator may want to dispense the first 200 droplets and then carry out a wash and aspiration process before dispensing the remaining droplets. The frequency of washing is largely determined empirically based on the behavior of the dispense liquid 320; however, the larger the number of dispenses the more frequent the wash cycles. Washing in an ultrasonic bath 130 to thoroughly clean crystal build up on the nozzle orifice 316 is recommended and significantly improves the dispensing positional accuracy and the consistency of the dispense droplet size. The robotic system 101 will automatically wash the nozzle 316 after a specified number of dispenses dependent upon the composition of the dispense solution, which is configurable by the operator. To maintain the system, washing is needed before and after each liquid handling cycle.
Some dispense solutions contain very high concentrated salt in volatile solvents, which may rapidly form crystals at the tip of the nozzle orifice 316. In one embodiment, a sapphire orifice of 75 micrometer inside diameter is used and can be easily clogged by build up of the salt crystals. To help prevent the nozzle 316 from clogging, high dispensing speed may be provided. Two significant advantages come with the high speed dispense. First, the high speed dispense can spot large amount of droplets before clogging happens. Second, high speed dispenses make faster liquid flow that slows down the crystal build up at the tip of the nozzle orifice.
However, the high speed dispense requires the robot 101 to move quickly and precisely to all positions the operator selects as the dispense pattern. In one embodiment, the Cartesian robot 101 with linear servomotor and distributed closed-loop proportional-integral-derivative (PID) motion control system meets these requirements. A dispense speed higher than 2 Hz, for example, and in some cases higher than 10 Hz, may be achieved.
After the liquid handling process is completed, the imaging device with illumination can be used to obtain images of the dispense pattern. From these images, post process inspection, quality control, and verification of liquid handling can be accessed. Also, the liquid handling results can be stored using image files. Using a specific defined data format (for example, XML data format), the actual dispensed pattern information can be saved automatically. The dispensed pattern file can be used by another instrument for further analysis or action (e.g., mass spectrometry). Thus, the system may export the coordinates for dispensing as determined by the operator prior to dispensing and/or export the actual coordinates of the dispensed material as determined through post dispense image analysis. A comparison of the differences can be used to improve dispense protocols and to formulate well behaved materials used in dispensing.
After completion of liquid handling for all of the objects, the operator or robot software may execute a final wash and clean up process to maintain the system in good working order.
The hardware configuration, electronic and software control, and methodology for aspiration, dispensing and in-process quality control of the present invention have been calculated to result in reliable dispenses for less than approximately 10 nanoliters, and more particularly, less than approximately 5 nanoliters, for example, from approximately 2 to 2.5 up to 4 nanoliters. The methods and apparatus of the present invention are believed to be capable of dispensing down to approximately 1 nanoliter or less. Such volumes reflect a 10-20 fold improvement over existing solenoid based dispensers.
Specific embodiments of an invention are described herein. One of ordinary skill in the art of small volume liquid handling systems will recognize that the invention has other applications in other environments. In fact, many embodiments and implementations are possible. For example, the present invention could be applied to applications like high throughput compound screening, and DNA, protein and cell array studies. In particular, this invention may be applied in proteomics and drug study applications in which protein or small molecule drug (and metabolites) spatial distribution information is profiled using mass spectrometric imaging techniques. Proteomics involves the identification and characterization of cellular proteins and the determination of their role in cell function, disease and response to exogenous or external factors and stimuli such as drugs, toxicants, and stress.
Another use of this invention is to enable the identification of proteins, protein structures, interactions and pathways so that new disease markers and drug targets can be identified that will help create new products to prevent, diagnose and treat disease. A further use of this invention is to enable the determination of the spatial accumulation of small molecule drugs and their metabolites in target tissues or their relative distribution in diseased versus normal tissues using mass spectrometry analysis.
The recitation “means for” is intended to evoke a means-plus-function reading of an element in a claim, whereas, any elements that do not specifically use the recitation “means for,” are not intended to be read as means-plus-function elements, even if they otherwise include the word “means.” The following claims are in no way intended to limit the scope of the invention to the specific embodiments described.
This application claims priority to U.S. Provisional Application No. 60/781,932 filed on Mar. 13, 2006 and U.S. Provisional Application No. 60/811,319 filed on Jun. 6, 2006, the contents of which are hereby incorporated herein by reference for all purposes.
| Number | Date | Country | |
|---|---|---|---|
| 60781932 | Mar 2006 | US | |
| 60811319 | Jun 2006 | US |
