Obtaining value from the human genome: a challenge for the pharmaceutical industry

Are grammatical representations useful for learning from biological sequence data?--a case study

This paper investigates whether Chomsky-like grammar representations are useful for learning cost-effective, comprehensible predictors of members of biological sequence families. The Inductive Logic Programming (ILP) Bayesian approach to learning from positive examples is used to generate a grammar for recognising a class of proteins known as human neuropeptide precursors (NPPs). Collectively, five of the co-authors of this paper, have extensive expertise on NPPs and general bioinformatics methods. Their motivation for generating a NPP grammar was that none of the existing bioinformatics methods could provide sufficient cost-savings during the search for new NPPs. Prior to this project experienced specialists at SmithKline Beecham had tried for many months to hand-code such a grammar but without success. Our best predictor makes the search for novel NPPs more than 100 times more efficient than randomly selecting proteins for synthesis and testing them for biological activity. As far as these authors are aware, this is both the first biological grammar learnt using ILP and the first real-world scientific application of the ILP Bayesian approach to learning from positive examples. A group of features is derived from this grammar. Other groups of features of NPPs are derived using other learning strategies. Amalgams of these groups are formed. A recognition model is generated for each amalgam using C4.5 and C4.5rules and its performance is measured using both predictive accuracy and a new cost function, Relative Advantage (RA). The highest RA was achieved by a model which includes grammar-derived features. This RA is significantly higher than the best RA achieved without the use of the grammar-derived features. Predictive accuracy is not a good measure of performance for this domain because it does not discriminate well between NPP recognition models: despite covering varying numbers of (the rare) positives, all the models are awarded a similar (high) score by predictive accuracy because they all exclude most of the abundant negatives.

Transgenic knockouts as part of high-throughput, evidence-based target selection and validation strategies

The worldwide genome sequencing projects are helping to define the size and complexity of the expressed genome and are thereby identifying an unprecedented number of genes of uncertain disease alignment and unknown function. It is widely recognized that, within the pharmaceutical industry, a significant commercial advantage will accrue to those companies that most effectively gather and integrate additional biological information into their therapeutic target selection and drug progression strategies. This article presents the rationale for including comparative phenotypic information obtained from transgenic gene knockouts as an integral part of any future therapeutic target selection strategy.

Applications of developmental biology to medicine and animal agriculture

With the complete sequence of the human genome expected by winter 2001, genomic-based drug discovery efforts of the pharmaceutical industry are focusing on finding the relatively few therapeutically useful genes from among the total gene set. Methods to rapidly elucidate gene function will have increasing value in these investigations. The use of model organisms in functional genomics has begun to be recognized and exploited and is one example of the emerging use of the tools of developmental biology in recent drug discovery efforts. The use of protein products expressed during embryo-genesis and the use of certain pluripotent cell populations (stem cells) as candidate therapeutics are other applications of developmental biology to the treatment of human diseases. These agents may be used to repair damaged or diseased tissues by inducing or directing developmental programs that recapitulate embryonic processes to replace specialized cells. The activation or silencing of embryonic genes in the disease state, particularly those encoding transcription factors, is another avenue of exploitation. Finally, the direct drug-induced manipulation of embryonic development is a unique application of developmental biology in animal agriculture.

Genetic screens in yeast to identify mammalian nonreceptor modulators of G-protein signaling

We describe genetic screens in Saccharomyces cerevisiae designed to identify mammalian nonreceptor modulators of G-protein signaling pathways. Strains lacking a pheromone-responsive G-protein coupled receptor and expressing a mammalian-yeast Galpha hybrid protein were made conditional for growth upon either pheromone pathway activation (activator screen) or pheromone pathway inactivation (inhibitor screen). Mammalian cDNAs that conferred plasmid-dependent growth under restrictive conditions were identified. One of the cDNAs identified from the activator screen, a human Ras-related G protein that we term AGS1 (for activator of G-protein signaling), appears to function by facilitating guanosine triphosphate (GTP) exchange on the heterotrimeric Galpha. A cDNA product identified from the inhibitor screen encodes a previously identified regulator of G-protein signaling, human RGS5.