Skip to main content
Premium Trial:

Request an Annual Quote

Spot-Watching: VigeneTech Takes a Tailored Approach to Image-Analysis Software

Premium
VigeneTech, a small software firm based in Carlisle, Mass., is building a loyal base of customers and instrumentation partners for its tailored approach to image analysis.
 
Over the last two months, the firm, which was founded in 2003 and employs just over a dozen people, made headway in its strategy to expand its network of original equipment-manufacturing partners with some high-profile life-science vendors.
 
In September, VigeneTech and Millipore launched a software package called Milliplex Analyst, which VigeneTech developed exclusively for Millipore to analyze multiplex assay data from the Luminex platform.
 
Soon afterward, VigeneTech and GE Healthcare’s Whatman subsidiary launched a customized version of VigeneTech’s MicroVigene array-analysis package to analyze data from Whatman’s FastQuant protein arrays.
 
And in mid-October, VigeneTech began an OEM partnership with Chinese microarray developer CapitalBio to develop a customized version of MicroVigene that will be integrated with CapitalBio’s LuxScan microarray scanners. That collaboration has led to a product that is now being launched, CEO Minzi Ruan told BioInform in an email.
 
VigeneTech, which also has offices in Hangzhou, China; Tokyo; and Paris, relies on an open architecture that “more easily provides customized solutions,” compared to other image analysis platforms, Ruan said.
 
Ruan said that her team, which includes six image-analysis software developers, works closely with its customers in academia and industry to take the same customized approach for its off-the-shelf software products as it does in its OEM software agreements.
 
At the core of VigeneTech’s software is a suite of “unique and advanced image analysis algorithms” for spot detection, splitting, and grid placement that are “by far the best in any commercial image analysis software that we know of,” Ruan said.
 
Compared to Definiens’ technology “our spot detection and splitting are much [more] robust and accurate,” she said. “Also our software is much more user-friendly and automated. Quality, performance, ease of use, flexibility, and extensibility are the five main areas that we have really focused on and that makes our software so superior,” she said. 
 
The firm’s core platform is the MicroVigene array-analysis suite, now in version 3.0, which was originally developed for DNA microarray analysis and later expanded to support protein and other microarray-based image analysis.
 
“We have a full line of customized solutions for protein microarray platforms in functional expression array, semi-qualitative and quantitative multiple well protein/antibody arrays, and reverse-phase protein arrays for biomarker discovery, drug validation, clinical research, and diagnostics,” Ruan said.
 
The firm has recently expanded the same image-analysis engine to “challenging images in cell/tissue-based applications, where our technology shows much superior power in quality and automation,” Ruan said. The company launched the first product of this effort, CellVigene, designed for analyzing high-content screening images and tissue microarrays, last August.
 
“We would like to use the same approach as protein arrays to develop a whole line of customized solutions for cell/tissue-based image applications,” she added.
 
Several VigeneTech customers cited a key benefit the software’s customization capabilities. For example, BioReliance, which is using the software for its animal pathogen-screening service, worked with the company to develop a microarray analysis package tailored to its needs.
 
“With VigeneTech we were able to take their existing commercial array software and customize it to our specifications,” Tibyron Quinn, a research scientist and project manager at BioReliance, told BioInform in an e-mail. “I went back and forth with the development team for many months until we got exactly what we needed.”
 
BioReliance recently upgraded to a system capable of screening multiple pathogens simultaneously and wanted to be able to analyze “massive amounts of array data,” and take the “intense legwork” out of the testing process, he said.
 
He explained that his firm worked with VigeneTech to automate the same steps a scientist would take to compare data and then the computer could “crunch through all of the calculations it takes to make a final determination.”
 
“VigeneTech provided us with such a complete package that we are able to generate a final report, which goes to the customer directly from the software,” he said. “I have used a lot of other array programs in the past but none which you could have customized to such a high level.” 
 
Where is the Grid?
 
Joshua LaBaer, director of the Institute of Proteomics at Harvard Medical School, told BioInform that MicroVigene’s automation capabilities — particularly for spot-finding — have been a helpful feature in analyzing his lab’s nucleic acid programmable protein arrays, or NAPPAs, high-density arrays that do not require researchers to first purify proteins.
 
One challenge in protein array analysis, he said, is that a printer might only theoretically put spots in exactly the right position. “When you try to put a grid over what you printed, it won’t exactly line up.”
 
While many software applications “allow you to stretch and pull and move and push until the thing lines up the way you want it to,” he said. “If you’re going to run 100 samples, that becomes a real pain, because it means you have to sit there and yank and pull and tug for every one of the 100 samples, and you don’t have to do that with this software.”
 

“But if you’re going to run 100 samples, that becomes a real pain, because it means you have to sit there and yank and pull and tug for every one of the 100 samples, and you don’t have to do that with this software.”

For labs with several thousand spots on a slide and hundreds of slides “it’s a nightmare,” LaBaer said. “We had to do this all manually until we got this software, and it really fixed the problem for us.”
 
The software also aids in quantification, he said. “You may still have to go back statistically and correct for batch results, but that you would have had to do anyway. The difference is getting the information off your slides into a quantitative format — that is the thing that is tricky. And that was a pretty big advantage for us,” LaBaer said.
 
Eugenie Hainsworth, senior engineer in LaBaer’s lab told BioInform in an email that MicroVigene “does fantastically well” setting the “proper grid” and locating spots “when almost all the spots on the array are missing,” which can happen “when you are looking for an interaction that only occurs on a few specific spots on your array.”
 
LaBaer said that even though he has bioinformaticists in his lab, they didn’t venture to write software for spot finding. “There is no sense in reinventing the wheel,” he said.
 
“We tried a lot of software and nothing compared,” he said, adding he did not wish to name the software he and his colleagues tried. MicroVigene “approaches each spot from multiple sides and it looks at signal intensity until it sees a rise in signal intensity and says, ‘Ah, I have found the beginning of a spot right here.’”
 
LaBaer added that the software allows users to see “where your peaks and valleys are” — a characteristic that was “particularly useful” in one instance in which the array chemistries were causing spill-over signal between spots. In a three-dimensional rendering, the scientists found it easier to discern spill-over by using the software.
 
“It looked as if there were all mini-mountains, but they were all on the side of the spot facing the other peak. You could tell it was spill-over, it wasn’t from the center of the peak, it was from the edge,” he said.
 
Hainsworth said that VigeneTech let the scientists use a demo version of the software to try with various arrays and compare it with other array-analysis programs. “We had several researchers here doing different experiments, which generated wildly different types of array images,” she said.
 
VigeneTech incorporated features specific to arrays in LaBaer’s lab into the software and gave the lab a version of the software that incorporated features other users had requested as well, Hainsworth said.
 
LaBaer said that he and Hainsworth spent “a lot of time walking through things we wanted” with VigeneTech and found the firm “incredibly” responsive.
 
“They would go away and fix it and come back and say ‘Try this,’” he said.
 
Making the software work better for the lab’s routine tasks, such as being able to work through many slides at a clip, took a few months, LaBaer said, adding that “it changed our lives, it was a quantum leap above what we had [been using.]”
 
Reversing Phase
 
Emanuel Petricoin, co-director of George Mason University’s Center for Applied Proteomics and Molecular Medicine, and his colleagues used the VigeneTech software to address another type of image-analysis challenge related to protein arrays.
 
George Mason’s reverse-phase protein array, or RPPA, measures proteins in clinical samples such as biopsies to measure, for example, tumor antigens.
 
“It wasn’t like there was some simple software imaging solution for these,” he told BioInform. “We almost needed customized software.”
 
A particular challenge for the RPPA is background subtraction, Petricoin said. “Since we’re printing usually cell lysates, there is negative background and you have to do subtraction,” he said. Users need to “assure you are in the linear dynamic range of the assay.”
 
Petricoin said that he and his colleague Lance Liotta discussed the image analysis challenges with Ruan and she and her colleagues adapted MicroVigene to incorporate their feedback.
 
MicroVigene had a tool “that would automatically determine the linear slope of the assay and they incorporated more tools based on our advice into that,” he said. “They were very focused on helping us,” he said.
 
Ruan studied at the University of Science and Technology in China, received her PhD in Physics from Case Western Reserve University and did a post-doctoral fellowship in mass spectrometry at North Carolina State University.
 
She held R&D positions at several companies prior to founding VigeneTech. “I had managed several large-scale projects in network management while I was with International Network Services [a subsidiary of British Telecommunications] and Lucent before [I] started VigeneTech,” she said. “My management experience with the Chinese Academy of Sciences and Lucent prepared me for my entrepreneurial venture.”

Filed under

The Scan

Panel Votes for COVID-19 Pill

A US Food and Drug Administration panel has voted to support the emergency use authorization of an antiviral pill for COVID-19 from Merck and Ridgeback Biotherapeutics, CNN says.

But Not Harm

New Scientist reports that UK bioethicists say that though gene editing may improve food production, it should not harm livestock welfare.

Effectiveness Drop Anticipated

Moderna's Stéphane Bancel predicts that that current SARS-CoV-2 vaccines may be less effective against the Omicron variant, the Financial Times reports.

Cell Studies of Human Chromatin Accessibility, SARS-CoV-2 Variants, Cell Signaling Networks

In Cell this week: chromatin accessibility maps of adult human tissues, modeling to track SARS-CoV-2 variants of concern, and more.