The Introduction And The Application Of Hub's Rule In Gene-Oriented Bibliome

Mining the biology literature in bulk could be another high-throughput strategy ofbio-information retrieving along with the transcriptomics and the proteomics inpost-genomic era, which had been defined asthe bibliomics. In protein-protein interactionnetwork, there's a dogma that demonstrates a positive correlation between lethality andconnectivity of a node, also known as the centrality-lethality rule. This research discussedthe existence and the basic principles of a centrality-lethality-like rule (hub's rule) ingene-gene co-occurrence network (GGCON) of bibliomics.Genes with high connectivity in GGCON tend to have more chance to be emphasized byscientific community if they have not. This rule was confirmed by experiments, and theavailability of its application in highlighting or prediction of topic-related genes wasdeduced through modeling with the principle of conceptual biology. In this modeling, thecore genes and the extended genes were discriminated depending on their relationship tothe topic. The transformation of the extended gene to the core gene was determined by twoimpact factors (green and red), which were discriminated by their dependence on theconceptual chains.. The impact of the red factor was measured and simulated through linearregression and programming. It was confirmed that the extraction of topic-related genesthrough hub's rule was reasonable, continuable and practicable. Genes related to Alzheimerdisease (AD), breast neoplasm (BN) and other human diseases were extracted in the end ofthis chapter.To search for topic-related genes from a noisy dataset, the GeneRankV1 algorism wasdeveloped which implemented the hub's rule. It made a gene highly ranked if it was linkedby other highly ranked ones. A golden standard of AD related genes listed by an AD expertwas chosen to measure the performance of the algorism. Genes in this golden standard weresuccessfully selected out from a noisy dataset when applying the algorism. The true positive rate of the top 15% genes from the list was much higher than that of Entrez Geneand Gene Cards. Then, the Gengle platform, which implemented the GeneRankV1 algorism,was constructed. It had covered 1131 human diseases, 199 pathways, 10699 human genes,96547 MEDLINE entries and 9437 gene patents by the day of 2006-6-30. The platformorganized information in a gene-oriented way, ranked the topic-related gene robustly andprovided the literature of the relationship between genes and topics. The patented genescould also be extracted from noisy gene patents by the construction of patented-gene-omemodule of Gengle.The selection of the analgesic targets is somewhat similar to the search of diseaserelated genes. Known and potential analgesic targets existed in the pain related MEDLINEentries contaminated by noises. So the GeneRankV2 algorism was introduced to 83450pain related MEDLINE entries. The algodsm had selected out the known analgesic targets,extracted the ignored ones and predicted the potential ones. Considering the ignored oneswere less developed and covered by less intellectual property (IP) barrier, 11 ignoredanalgesic targets were selected for further investigation by GeneRankV2. The developingstrategy of analgesics was finally determined with nicotinic acetycholine receptor (nAChr)as the drug target and the prokaryote expression as the preparation method. Anexpression-purification system was developed for the preparation of the ligand of nAChr,which integrated the renaturing, enterokinase digestion and the purification on oneNi²⁺-affinity chromatography column.This work has confirmed the availability of the hub's rule-based extraction oftopic-related genes theoretically and experimentally. It is expected that Gengle will becomean important supplementary to the current gene-oriented information platform. The firstNational Bio-sequence database of Chinese patent (NASDAP) can give comprehensiveconsults on drawing IP strategies in the areas of both pharmaceutics and diagnostics. Thehub's rule-based analgesic target selection as well as the construction of theexpression-purification system will facilitate the subsequent works in analgesicsdeveloping.