Open Genomics
Private companies that use open science / open source approaches in genomics
Via Josef Jackson:
- DNAnexus was founded out of Stanford a few years ago and Google just invested in them. They are hosted the Short Read Archive which was de funded by the government in the US so the company has taken on the hosting duties for it and remains committed to keeping it open access.
- Numedii, a company that searches public genomics data to find uses for drugs that can be repurposed.....(approved for some condition but can actually treat something else we wouldn't expect until we data mine the relationships).
- NextBio is another commercial bioinformatics company that launched a free to use version for academics.
Commentary
By Lise Lévesque in the Cooperation Commons mailing list:
"There is an urgent call for openness and sharing among the genomics scientists in general. However, this means different things depending on the research field.
Functional genomics, which is looking at variations in molecular patterns, needs to use systems biology and is analysing the differences in the whole molecular composition of different samples. Databases containing molecular patterns that were characterised by other researchers need to be comparisions. Comparisions can themselves become important data to be stored and made accessible. However, even if the need for pooling results is obvious, there are many non-scientific factors that may influence the researcher's openness. First, there is a need for harmonisation of ontology. Networks of researchers are working at setting standards for sharing that type of information. But there is also the question of when would a researcher want to share the data. It could make sense to share the data once the results have been published.
More complex questions may arise in the field of genetic epidemiology, where personal genetic, medical and social data from tens of thousands of individuals are needed to generate results that will be statistically significant. So, yes, data sharing is required. There are complex national and international regulations about personal data exchange across borders. Harmonising the methods also is required. There are many challenges. Building databases and biobanks is costly and takes time. Researchers are getting organised in networks to address these questions. They are now cooperating to build these databases and biobanks and set a common language to use them. But when time will come to pool everyone's work together, how should that be done? Should they pool the data in a central repository? There will be technological answers to that question, and they need to be heard. But how will the "human factor" influence the process? Ethics and laws may have an influence on what can (ethically and legally) be done. Quality control will be needed: how will it be done? Maybe researchers will rather exchange data directly, selecting the appropriate partner within large collaborative networks. What are the features of the "appropriate" partner? So, it is more complex than simply posting a new information on the web."