In the internet age, all scientific data-- raw data and the algorithms used to analyze it-- need to publicly available without restriction. If it is publicly funded, the data belongs to the public. If it is published, journals have an obligation to provide readers who wish to review the data with ample opportunity to do so.
Decisions that change our lives are made based on "what science says". Claims based on "private" data aren't science, and the appropriate inference drawn from scientists who resist full public disclosure of data is that they are lying.
Scientific data needs to be treated legally like financial data of public corporations.
Transparency in science is essential, and should be required by law.
It will be, hopefully soon. It will be a revolution in science, and you'll be surprised how many scientific claims diverge considerably from the data on which they are ostensibly based.