Sir

As representatives of the funding agencies that support the large-scale genome sequencing of parasites and related genomic research, we disagree with Colin Macilwain's description in the News report “Biologists challenge sequencers on parasite genome publication” (Nature 405, 601–602; 2000) of the spirit of collaboration among the sequencers, funders and research scientists involved.

The data-release policy of these genome projects is for each sequencing centre to release unfinished sequence and assembled ‘contigs’ to a publicly accessible website. The policy encourages investigators to use the sequence data for gene-by-gene discovery. Already, in the case of the malaria parasite, more than a dozen peer-reviewed publications have used the preliminary sequence data and acknowledge the source of the sequences.

The sequencing centres and their collaborators need to keep responsibility for the computational analysis of unpublished sequence data on a chromosome or genome scale, for various reasons. First, the data are preliminary, and analyses will need correcting or completing as new data are added and existing data are modified. Second, the sequencing investigators and their collaborators should be allowed to confirm their research and publish the initial analysis of the data they have worked hard to generate.

Unfortunately, a few researchers view scientists at sequencing centres as service providers. In reality, they are conducting cutting-edge research, such as developing strategies for dealing with genomes with unusual base composition (Plasmodium falciparum is 80% A–T), including new methods to close difficult gaps in the sequence. In addition, they have designed and modified computer algorithms to identify and more accurately predict parasite genes and intron–exon boundaries.

Sequence data are no different from other types of biological data, and their generators deserve the same rights — to do an initial analysis and to publish the results — that other researchers have traditionally enjoyed. These data and their analysis should be published in peer-reviewed journals as quickly as possible. Until then, the sequencing centres and their collaborators are seeking ways to provide added value to the unpublished sequence data.

At a meeting of the malaria genome project in June at the Wellcome Trust Genome Campus in Hinxton, UK, the sequencing centres agreed that a first-pass annotation would be added to their websites, including posting the results of blast hits for all the open reading frames. Participants in the malaria genome project have agreed on the development of the PlasmoDB database ( http://plasmodiumdb.cis.upenn.edu), a web-accessible relational database facilitating the analysis of various genome-related data. With support from a grant from the Burroughs Wellcome fund, this database is being developed by David Roos and his colleagues at the University of Pennsylvania and by Ross Coppell at Monash University.

Contrary to your News report, the PlasmoDB team is not supported for, or engaged in, any effort to annotate the genome. Discussion is under way, however, to determine the extent of automated annotation that will be provided through PlasmoDB for unfinished sequence available from the sequencing centres.

WHO's programme for research and training in tropical diseases (see below) is to provide support for a ‘help desk’, in addition to the current distribution of data through a CD-ROM version of Malaria Information Resource, to help investigators who lack the computational resources to access and analyse the data.