The development of computer technology and the Human Genome Project has led to an emerging discipline, bioinformatics, which includes work in two cross-cutting areas: used to establish the information system framework required by modern biology Information management systems, analytical tools, and communication networks), that is, bioinformatics in the traditional sense; computation-based research work aimed at understanding basic biological problems, that is, computational biology . The Bioinformatics and Genome Research (Bioinformatics and Genome Research) series of conferences began in 1990, and the 6th International Bioinformatics and Genome Research Annual Conference was held in San Francisco, California, USA from June 11 to 12, 1997. The main topics include emerging new technologies, functional analysis of genes, new data tools and pharmaceutical-led gene and protein discovery [1]. The relevant contents are briefly introduced as follows:
1. Emerging technologies
Klingler (Lncyte pharmaceuticals, PaloAlto, CA, USA) emphasized that genomics is pushing the pharmaceutical industry into the information age. With the increasing generation of sequence, expression, and mapping data, information tools to describe and develop these data have become critical to the task of genomic research. He talked about Incyte pharmaceuticals' contribution to large-scale genomic data and bioinformatics.
Lipshutz (Affymetrix, Santa clara, CA, USA) describes a method for genomic research using DNA probe arrays. The principle is that the human genome can be achieved through more efficient methods of mapping, expression detection and polymorphism screening. Sequencing. The light-mediated chemical synthesis method is used to manufacture miniaturized arrays of high-density oligonucleotide probes. This array of oligonucleotide probes designed through software packages can be used for polymorphism screening and gene analysis. Type and expression detection. These arrays can then be directly used for parallel DNA hybridization analysis to obtain sequence, expression, and genotyping information. Milosavljevic (CuraGen, Branford, CT, USA) introduced a new gene expression detection system based on a dedicated quantitative expression analysis method, and a gene discovery system GeneScape. In order to effectively sample expressions, a fragment pattern was deliberately made to understand the occurrence and redundancy of subsequences of specific genes. He verified the performance of the technology in large-scale studies of differential gene expression in yeast, and discussed the application of the technology in the basic research of gene expression, biological function and disease.
2. Functional analysis of genes
Overton (University of Pennsylvania School of Medicine, Philadelphia, PA, USA) discusses the next stage of the Human Genome Project—genome-level analysis of gene function. The analysis, management and visibility of the data generated in this stage will undoubtedly be more complicated than in the first stage. He introduced E-poDB, a prototype system for functional analysis of the erythropoiesis of vertebrate hematopoietic systems, which includes the Kleisli system for integrating data resources and the bioWidget graphical user interface for establishing visualization tools on the internet or intranet. EpoDB may guide experimenters to discover new drug targets for erythroid development that are not possible with traditional experimental methods. The pharmaceutical industry is interested in new drug targets. EpoDB provides such an opportunity, which may be its most exciting The place.
Sali (Rockefeller university, New York, NY, USA) discusses the structural modeling of homologous proteins. Comparative protein modeling (comparative protein modeling) is also called homology modeling (homology modeling), that is, using the protein structure determined by experiment as a model (model) to predict the conformation of another protein (target) with a similar amino acid sequence. This method is now sufficiently accurate and is considered to work well, because a small change in protein sequence usually only results in a small change in its three-dimensional structure.
Babbitt (University of California, San Francisco, CA, USA) discusses methods for identifying distant proteins through database search. Understanding the interdependence of the structure and function of the protein superfamily requires understanding the implicit limitations of a particular structural template that nature shapes. The most interesting relationships between protein structures are often expressed in divergent sequences, so it is important to distinguish between low-scoring sequences with significant biological relationships and sequences with high scores but less significant biological relationships. Babbit proved that by using BLAST search, it is possible to identify distant relationships in the low-score areas obtained from database searches. Levitt (Stanford univeersity, Palo Alto, CA, USA) discussed protein structure prediction and a method for automatically modeling functions from sequence data only. Gene function depends on the tertiary structure of the protein encoded by the gene, but the number of protein sequences in the database doubles every 18 months. In order to determine the function of these sequences, the structure must be determined. Homology modeling and ab initio folding are two existing complementary protein structure prediction methods; homology modeling is accomplished through segment matching. The computer program abandoned SegMod based on Homology modeling method.
3. New data tools
Letovsky (Johns hopkins University, Baltimore, MD, USA) introduced the GDB database, which is composed of many different maps of each human chromosome, including the contents of cytogenetics, genetics, radiation hybridization, and sequence tag sites (STS), And the maps obtained by different researchers using the same method. As far as location query is concerned, it is useful to be able to search all maps, regardless of their type and source, or whether they happen to contain markers to approve areas of interest. For this purpose, the database uses a common coordinate system to arrange these maps. The database also provides a high-resolution map that shares many signs with other maps as a standard. The correspondence between the targets of the shared logo allows the assignment of standard atlases that are equal to all other atlases.
Markowitz (Lawrence berkeley Laboratory, Berkeley, CA, USA) discussed the relationship between distributed databases and local management, and the development of molecular biology databases (MDBs) using tool-based methods. Many solutions are currently promoting the search for data from multiple sources of MDBs, including the establishment of data warehouses; this requires a holistic view of the combination of various MDBs and loading data from member MDBs into the central database. The main problem with these solutions is the complexity of developing global views, building a huge data warehouse and synchronizing the integrated database with the evolving member MDBs. Markowitz also discussed the object protocol model (OPM) and introduced tools that support the following purposes: creating OPM views for text files or relational MDBs; making MDBs into a database directory, providing MDB names, positioning, and topics , Access to information and links between MDBs and other information; explain, process and explain multi-database queries. Karp (SRI international, Menlo Park, CA, USA) explained Ocelot, an object-oriented knowledge presentation system (an artificial intelligence version of an object-oriented system) that can meet the needs of managing biological information. Ocelot supports schema evolution and uses a new optimized parallel control mechanism (simultaneous multiple data access processes). Its sketch-driven graphic editor provides interactive browsing and editing functions, and its annotation system supports the database Structural communication between developers.
While discussing the function of E. coli protein, Riley (Marine biological Laboratory, Woods Hole, MA, USA) specifically mentioned the GPEC database, which includes information on the function of all E. coli genes determined by experiments. The largest proportion of proteins in this database are enzymes, followed by transport and regulatory proteins.
Candlin (PE applied Biosystems, Foster City, CA, USA) introduces a new relational database system BioLIMS that stores data directly from the ABIPrism dNA sequencer. The system can be integrated with the data of other sequencers, and can be automatically invoked with other software packages. It provides an open and extensible bioinformatics platform for the integration of sequencers and sequence data.
Glynais (NetGenics, Cleveland, OH, USA) believes that one of the most critical problems in bioinformatics is the lack of flexibility of software tools and databases. However, the development of software technology has been borrowed from the development experience of other fields such as the financial industry and the manufacturing industry, which can enable software from different software vendors running on various hardware systems to work together. The international standard for this system is CORBA, a software system developed by more than 250 major software and hardware companies. The joint use of CORBA and Java can develop a variety of network applications that access any kind of data or software tools through a common user interface, including bioinformatics applications. Overton disagreed with Glynias's idea. He emphasized that CORBA is only useful for software integration. Incompatible database software may be the most difficult problem facing computational biology. Some pharmaceutical companies and database warehouses recently funded a link with OCRBA Different database plans [2,3].
Fourth, the discovery of pharmaceutical pioneers
Burgess (Sturctural bioinformatics, San Diego, CA, USA) discusses computational issues in protein structures that bridge the gap between genomics and drug design. In the absence of accurate description data of major disease genes or drug targets, drug designers have to adopt large-scale expression protein screening methods; while structural bioinformatics uses a more practical and effective calculation method directly from sequence data In order to determine the fine structural characteristics of the active site of the target protein, it uses an integrated expert system for rapid computational screening from real or virtual chemical libraries, which can reach a large scale.
Elliston (Gene logic, Columbia, MD, USA) discusses the process of discovering new molecular targets in the development of therapeutic drugs, focusing on gene discovery methods. He believes that with the approaching sequencing of the human genome, the characteristics of almost all genes will be revealed at the sequence level. However, the understanding of genes will depend on more information than just sequences. The first type of information to be considered is the transcriptional expression level information. GeneLogic ’s GeneExpress is a combination of mRNA expression profiles, transcription factor sites, A database of new genes and expressed sequence tags.
Liebman (Vysis, Downess grove, IL, USA) introduced the computational and experimental methods developed by Vysis, which are not only used to manage sequence data, but also used for the following purposes: analysis of clinical databases and natural-mutation databases; development of new Algorithms to establish functional homology (different from sequence homology) to simulate biological pathways for risk assessment; target evaluation of drug design; link complex pathway characteristics to identify side effects; develop qualitative models of disease development and explain clinical consequences .
With the increasing number of new genes discovered, this question becomes particularly important: What is the function of the gene? Escobedo (Chiron technologies, Emeryville, CA, USA) proposed a method for this problem: combining functional clones of gene secreting proteins with screening these clones (possible drug targets). In this method, in vitro translation in the microsome cDNA library pool avoids labor-intensive cloning, expression, and purification steps. The translation products in the library pool are screened at the cellular level and tested for cell proliferation and differentiation. The role. For example, of the 111 clones identified by this method, 56 belong to known secreted proteins, 25 are membrane-associated proteins, and the other 30 have unknown functions and may be new proteins. A similar method is to construct a cDNA library of secreted proteins in a gene transfer vector transferred to a mouse model system to clone specific functional genes.
Ffuchs (Glaxo wellcome, Research Triangle Park, NC, USA) discusses the broader impact of bioinformatics: it not only affects the discovery of new drug target bases, but also greatly improves the status of pre-clinical and clinical phases of drug development importance. As we all know, no matter how careful the design of clinical trials involving thousands of patients (probably the most expensive part of drug development), it cannot select the right patient for the right drug. The method of dividing patient groups at the genomic level can greatly improve the efficiency of discovering new drugs. Fuchs introduced a system that combines patient genotype and phenotypic markers to improve the preclinical and clinical drug development process. He emphasized that genetics and bioinformatics data should be combined with chemistry, biochemistry, and pharmacology Integrated information management and analysis methods linked to medical data are extremely important.
Green (Human Genome Sciences, Rockville, MD, USA) introduced the data management tools used in his sequencing work. The challenge for EST-based sequencing methods is that after repeated sequencing of hundreds of cDNA clones, the resulting data piles up. Since most human genes are discovered by this method and sorted and organized in a database, the task of identifying open reading frames, overlapping maps of overlapping sequences, tissue-specific expression and low-abundance mRNA genes is daunting of. Human genome Sciences has developed some customizable database tools, which can include the following functions in the same database: accessing and retrieving data on WWW, sequence splicing, and research progress on potential drug target genes. These software tools that can manage multiple tasks—from annotating gene sequences to successfully developing gene products into the drug discovery process—are extremely promising to derive new drug targets from a genomic knowledge-based drug discovery method.
Summer-Smith (Base4 bioinformatics, Mississauga, Ontario, Canada) describes a related strategy. The tasks of the software tools required in the drug discovery phase are diverse, to be able to annotate genes and clarify their physiological and pathological functions and their commercial potential. The integration and analysis of information from such multiple sources can be easily accomplished in a derived, project-oriented database (PSD). Since the project runs through the entire process from discovery to development, and members who continuously join the background, PSD has become a key resource in project management and development.
According to Smith (Boston university, Boston, MA, USA) [2], we do not need a faster computer or more computer scientists, but more biologists and biochemists to explain the function of the sequence . This is a blow to some software or hardware experts, but the complexity of biological systems is daunting, and the understanding of gene function may require a combination of biological and computational methods. Exploring the function of genes is likely to take decades for biologists. This meeting showed that there is no single way to come up with an answer; however, combining computational biology with large-scale screening first identifies a chemistry A hit is a method of generating chemical tools to explore gene function. These chemical tools can then be used as "probes" to understand gene function. In the description of Butt (Gene Transcription Technologies, Philadelphia, PA, USA), this method is not only a simple method for checking gene function, but also a simple method for discovering chemical leads for potential drug targets. He described a method that can A large-scale yeast screening system to reconstruct the function of human genes in yeast. In this system, ligands can be quickly found in a chemical library. An important feature of this technology is that it is not only a screen for finding a ligand for a drug target, but also, because of the high speed of the system, it is also a screen for discovering a lead target gene. In the past, pharmaceutical companies in the world usually could only work on a limited number (about 20) of drug target genes at a time. In view of this, we need fundamentally different methods such as genomics to open the way to "new" Biological pathways. Due to advances in robotics and synthetic chemistry, the most critical issue in drug discovery is no longer to obtain a lead compound, but to target genes. This meeting is a good step for new biology developed from computational and experimental methods.
Automatic Filling Machine For Container 15-50L
Liquid Filling Machine,Automatic Can Filling Machine,Small Container Filling Machine,Automatic Filler
Changchun Rochiev Intelligent Equipment Manufacturing Co., Ltd. , https://www.rochiev-en.com