Drug developer Serono has expanded its use of Genedata’s Screener software to include high-content screening in addition to high-throughput screening, Genedata said this week.
The expanded collaboration helps validate Genedata’s strategy to pursue the data-analysis market for high-content screening — an effort that officially kicked off in late 2005 when it announced a co-marketing deal with HCS vendor Evotec Technologies.
The details of that collaboration called for the firms to combine Screener with Evotec’s EVOscreen and plate::explorer instruments and Acapella image-analysis software.
Then, in January of this year, Genedata released Screener 4.0, the first version of the software to include support for high-content screening data. And in November, BioInform’s sister publication Cell-Based Assay News reported that Genedata established an informal collaboration with Thermo Fisher unit Cellomics to integrate Screener with the image-analysis software that accompanies Cellomics’ ArrayScan and KineticScan high-content imaging platforms.
The firms said at the time that they were working with customers who wanted to integrate the Cellomics platform with Genedata’s software, but did not disclose the identities of any mutual customers.
This week, Genedata revealed that Serono was the driving force behind the Cellomics integration. Stephan Heyse, general manager of discovery informatics at Genedata, said that Genedata and Serono began collaborating in 2003 to deploy Screener as part of Serono’s HTS operations, and began expanding into high-content screening in 2005.
Heyse explained that the extended collaboration took advantage of several features available in Screener 4.0. In addition to the HCS capabilities, the software includes an application programming interface that enables customers to tightly integrate Screener with internal databases or data warehouses, LIMS, and other components of their informatics infrastructures.
Genedata worked with Serono to design an HCS analysis pipeline that linked Cellomics’ image-analysis software with an internally developed LIMS and Screener. The project automated a number of processes that had previously been done manually.
Heyse said that Genedata’s partnerships with Cellomics and Evotec underscore its “best of breed” approach in marketing its software. “Corporate IT infrastructures have become more complex and companies tend to design their own IT infrastructure and to buy different parts from different companies,” he said.
Genedata’s Screener is “very good in the algorithmic and visual aspects of high-throughput data quality control and algorithmic analysis, but we don’t do some things, like, for example, image management,” Heyse said.
Screener’s suite of APIs, therefore, enables users to link Screener to other software products that provide additional capabilities. As an example, Heyse noted that one API allows users to click on a well in a plate in Screener and pull up the corresponding image that may be stored in Cellomics Store or another database. “We have seen that there we need to provide an easy customizable link, and in other areas as well, to pull that data in,” he said.
Serono and Genedata co-presented a poster at the 2006 Society for Biomolecular Sciences conference in September outlining their data-analysis pipeline. According to the poster, which described a neurite outgrowth assay, Screener’s machine learning-based approach resulted in a better hit verification rate than that of traditional hit list generation methods based on a single-parameter threshold.
Officials from Serono could not be reached for further comment on the collaboration with Genedata.
Room to Grow
Heyse said that Genedata sees plenty of room for growth in the HCS data-analysis market. “I think there is definitely a trend to increase high-content screening,” he said. “If companies are doing more high-content screening — and that seems to be the case — then Screener will be used more in high-content screening too.”
Heyse said that most of Genedata’s customers are typically using HCS in one of two ways: either as a secondary screen following high-throughput screening with biochemical assays; or as a primary screen for certain biological functions “that are more [easily] measured at the physiological response level rather than the individual targets.”
As an example of the latter category, Heyse cited the neurite outgrowth study with Serono. “That’s a complex biological process with many targets involved, and if you just simply ask, ‘Does my compound inhibit or promote neurite outgrowth?’ then you can’t do a classic target-based screen. You need to look at the cells. And that’s where high-content screening comes in because there is no other method by which you can automate that procedure.”
Heyse said that most pharma companies are moving toward a combined approach of HTS and HCS. “The high-throughput screens, or the biochemical screens, once they’re set up are typically cheaper to run than high-content screens, but they provide less information for more targets,” he said. “They will be combined.”
Genedata hopes to benefit from this trend by ensuring that Screener meets the needs of both HTS and HCS users. Heyes noted that the firm faces competition from in-house developed systems as well as third-party visualization applications like Spotfire’s DecisionSite, but he noted that Screener offers the flexibility of “horizontal” systems like DecisionSite along with the specialized algorithms that are required for HCS data.
“Corporate IT infrastructures have become more complex and companies tend to design their own IT infrastructure and to buy different parts from different companies.”
Genedata also faces competition from a few HCS vendors that offer complete packages for HCS, including data-analysis software. Cellomics is one of these, but obviously sees some synergies with Genedata’s product. In addition, Molecular Devices this week launched new HCS software, including data-analysis software, which it is billing as an alternative to software such as Spotfire.
“Our focus is really the data analysis and providing a turnkey solution involving visualization, algorithmic processing, and data management,” he said. “That’s not providing the whole IT infrastructure necessarily, but excelling in this domain and connecting to other products that cover other domains better than we do.”
Heyes added that the Serono installation is typical of how most drug discovery companies are building their screening analysis pipelines as a “mosaic” of third-party and in-house packages. “They have Cellomics for the image analysis, they have their own LIMS systems, they have us for data analysis, and it forms an integrated workflow that is tuned to their needs because they have built the bridges,” he said.