Got any nifty apps for your lab?

It's an unfortunate truism that scientists often have better software tools for managing their music or family photos than they do for tracking their experiments and data. Major software companies tend to focus on much larger consumer and business markets, and offer little software for researchers.

As a result, others are rushing to occupy the niche, among them Digital Science, launched last week by Nature 's parent company, Macmillan Publishers. Digital Science's strategy is not only to develop its own products — independently and with partner companies — but also to tap into the innovation of researchers themselves. With few existing products available to satisfy their needs, a growing number of scientists have developed their own tools with which to better organize their research lives.

The company has launched an open call for researchers who have written promising software to submit proposals for turning it into a commercial product. Researchers understand their colleagues' needs best, but often "don't have resources to turn their software into a polished product", says Timo Hannay, Digital Science's managing director and former publishing director of nature.com. The company aims to partner with researchers, or their start-ups, to provide them with the financial, developer and business resources they need.

The goal is to offer researchers tools that are as intuitive and user-friendly as well-designed consumer software. The company will initially focus on text-mining software, metrics-based tools to help institutions and funders better assess the performance of their funding and researchers, and lab-management software to help to keep tabs on anything from experiments to reagents.

There is certainly a massive need for better software to increase productivity at every stage of the complex scientific workflow, says Alexander Griekspoor, who founded the company Mekentosj, based in Aalsmeer, the Netherlands, which produces software for molecular-biology applications. Sriram Kosuri, a bioengineer at the Wyss Institute for Biologically Inspired Engineering at Harvard University in Boston, Massachusetts, agrees. "Every bench scientist I know has their own, and self-admitted non-optimal way to organize their information," he says. Kosuri is a founder of OpenWetWare, a wiki for sharing lab protocols and data between biology groups worldwide. "I think there is an increasing willingness to pay for useful tools," he says.

Scientists struggling to organize thousands of article PDFs strewn across their hard drives have already embraced services such as Griekspoor's Papers, and London-based Mendeley, which bring the simplicity of iTunes to managing papers. Mendeley also offers social-networking facilities and other features.

Publishing giant Elsevier has recently entered the research-services market with its SciVal suite of metrics tools. It is one of several publishers that view providing institutions with performance-measuring applications as a strategic move, says David Bousfield, the London-based vice-president and lead analyst at Outsell, a publishing and information consultancy. Elsevier has also launched SciVerse, a platform for searching and sharing content from Elsevier's own databases and the web. The product provides programming interfaces that allow researchers to build their own customized applications on top of content from Elsevier and other data sources, such as government databases.

Kosuri thinks that Digital Science's strategy of refining software created by researchers makes sense. "Getting the most important sets of tools developed more extensively would be tremendous," he says. But Michael Eisen, a geneticist at the University of California, Berkeley, and co-founder of the Public Library of Science, notes that many researchers have little enthusiasm for devoting time to perfecting software they have written. "I expect few would be interested," he says.

"We're not trying to convince anyone to create a commercial product if the desire isn't already there," Hannay counters. "We're trying to tap into the small but significant proportion of researchers who have identified an unmet need and are trying to do something about it."

Freely available code isn't an ideal alternative — most open-source software for scientists "sucks", says Eisen, although "a lot of it is really good and essential to what we do". But many lab-management tasks can be done using generic consumer open-source tools, he says. "The people I know mostly use wikis to keep track of stuff in the lab: they're free, flexible, and easy to set up and use."

Eisen also warns against thinking of software as a panacea. "Everyone has aspirations to be better organized in the lab," he says, "they think there's a magic piece of software out there that will solve all their problems for them, but then they realize the problem is really that they're disorganized."

figure a