¹ú²ú͵ÅÄ

Program to Detect Palindrome Concentrations

in an RNA Sequence
 

Clemente Aguilar

Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas at El Paso

Abstract

Palindromes are especially interesting words in the genomic content due to their different biological roles: among other functions, they can serve as replication origins, or transcription sites in some viruses. We are developing a program to estimate the concentration of palindromes in a given sequence. The program is expected to be useful for counting concentrations of palindromes in a given RNA sequence and classifying them with a score system. The program is written in C, and has interfaces with EMBOSS and R statistical software.

Gene Therapy for Hemophilia
 

Ana Betancourt

Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas at El Paso

Abstract

Hemophilia A and B are X-linked bleeding disorders with Hemophilia A being the most common hereditary coagulation disorder. The severe form of Hemophilia is characterized by spontaneous bleeding of the joints and internal organs. Patients are being treated intravenously with plasma derived and recombinant FVIII or FIV (Hemophilia B), but some of the disadvantages of protein replacement therapy are the limited availability of purified proteins, its high cost, and the development of antibodies against FVIII or FIX. The development of successful gene therapy for Hemophilia would transform the life of patients facing abnormal bleeding and shortened life span by producing a stable amount of coagulation factor and eliminating the risk of infection with contaminated products, frequent IV injections, and reduced immunogenicity.

Bayesian Models for Identifying Changes in Gene Expression from Microarray Experiments
 

Huiqin Chen and Stephen Aley

Bioinformatics Program and Department of Biological Sciences

The ¹ú²ú͵ÅÄ of Texas at El Paso
 

Abstract

cDNA microarrays have been extensively and successfully used in many fields, such as basic scientific studies, drug-discovery, and diagnostic purposes. cDNA microarray data is characterized by a large amount of measurement, a high degree of measurement noise and variability and low replications. The objective of this project is to find a Bayesian framework for the analysis of DNA microarray expression data accommodating the typical characteristics of microarray data. Three Bayesian models including the non-informative prior model, the conjugate prior model and the hierarchical model have been proposed and tested by simulation. The data set for testing the Bayesian approach is the high density array experiment reported by Arfin et. al. which compared Escherichia coli cells that were wild type to cells that were mutant for the global regulatory protein Integration Host Factor (IHF). The main advantage of this data set is its four-fold replication for both wild type and mutant alleles. The results show that the non-informative prior model is not suitable for microarray data and the conjugate prior is a better model. The hierarchical model is best in erasing noise from the background.

The Protein Information Tool
 

Anurag Gautam

Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas at El Paso

Abstract

The Protein Information tool determines or deciphers the several properties of protein like amino acid composition, the molecular weight, extinction coefficient, hydrophobicity, nature of amino acids telling whether a particular amino acid in a given sequence is basic, acidic or neutral, Position of alpha Helix and Beta sheets regarding secondary structure based on Chou-Fasman algorithm. The tool asks the user to paste the particular protein sequence in a FASTA format and by selecting any one of the options depending on user’s choice of interest , it determines or deciphers the required property. The protein properties were deciphered accordingly to the code written separately for each property of protein in PERL (Practical Extraction and Reporting Language). For determination of properties of proteins some data was collected from Bioinformatics website like NCBI, SWISS-PROT, PDB etc. The data was used in order to do create a particular code for the determination of characteristic of a particular protein sequence. To make Protein Information tool more powerful and informative, several codes can be written in PERL to decipher other properties of protein sequence. The Perl codes can be combined with HTML (Hyper Text Markup Language) in order to make it available to user.

Confirmation of Computer Predicted Secondary Structure of

RNA 2 of NoV via Site Directed Mutagenesis
 

Kimberly S. Hogle,¹^,² John H. Upton,² Abel Licon,³ Ming-Ying Leung,¹^,4

and Kyle L. Johnson¹^,²

(1) Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas El Paso
(2) Department of Biological Sciences, The ¹ú²ú͵ÅÄ of Texas El Paso
(3) Department of Computer Science, The ¹ú²ú͵ÅÄ of Texas El Paso
(4) Department of Mathematical Sciences, The ¹ú²ú͵ÅÄ of Texas El Paso
 

Abstract

Nodaviruses are small (25-34 nm) icosahedral, single stranded viruses with 2 strands of positive sense RNA genome. We hypothesize that the RNA secondary structure at the 3’ end of NoV RNA2 is required for viral replication. Bioinformatic computer predictions of RNA secondary structure were analyzed to identify specific bases involved in pairing to form these structures at the 3’ end of NoV genomes. Construction of plasmids was carried out to include full length genomic sequence of NoV RNA sequences which are capable of transformation into competent cells and transfection into yeast. Site directed mutagenesis was performed to delete bases in NoV plasmid constructs that are involved in computer predicted structures. Plasmid sequencing was performed to confirm corresponding deletions and check for existence of other mutations in these purified plasmid preparations.

Developing a PostgreSQL Database for Hyper spectral Data Collected Using an Automated Robotic Cart
 

Kuldeep Matharasi,^1,2 Santonu Goswami,² John A. Gamon,³ and

Craig E. Tweedie²
 

(1) Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas at El Paso
(2) Systems Ecology Laboratory, Department of Biological Sciences, The ¹ú²ú͵ÅÄ of Texas at El Paso
(3) Center for Environmental Analysis (CEA-CREST) and Department of Biological Sciences, California State ¹ú²ú͵ÅÄ, Los Angeles, CA
 

Abstract

Ecologists and Environmental Scientists collect a huge amount of data from different field sites trying to find an answer to a research question. One of the main challenges faced by Ecologists and Environmental Scientists is the proper storage and use of data and sharing it for future use. These data hold the key for finding an answer to different research questions and also for proper monitoring of various environmental events.

Spectral data were collected as part of the Biocomplexity experiment in Barrow Alaska for 2005, 2006 and 2007 using a robotic cart and a hyper spectral spectrometer. Huge amounts of hyper spectral data were collected as part of the project for the three years. The cart system uses a hyper spectral spectrometer which collects optical data in 256 bands along three transects over a dry Arctic lake bed. Each of the transects is 300m long and optical data were collected three times a week for every meter of the tramline. Therefore in one day the cart system collects 300 optical data files which totals to about 900 optical data filed for the total length of 900m length for each day. Therefore, it collects about 2700 data files a week. So, on an average, it collects about 30,000 data files for a three month long season. This huge volume of data created strong challenges to handle data effectively for effective data processing because of the lack of a proper storage system. This challenge led us to design a database system using PostgreSQL which allowed us the ease of accessing the data effectively for the purpose of data processing and also to do the quality check of the data at a minimal time.
 

Identification of Determinants of Prevalence of

O-glycosylation Sites in a Protein Sequence
 

Deepthi P Matta and Ming-Ying Leung

Bioinformatics Program and Department of Mathematical Sciences

The ¹ú²ú͵ÅÄ of Texas at El Paso

Abstract

O-glycosylation is one of the most important, frequent and complex post –translational modifications in proteins. Glycosylation affects many protein critical functions including cellular communication, half –life and structure (Jenkins and James 1980). Glycosylation also plays an important role in pathologies of some diseases and the altered glycosylation is implicated in cancer, mucosal diseases and pathogen–host interactions. O-glycosylation is the most common post-translational modification which occurs at Serine and Threonine sites in an amino acid sequence. O-glycosylation is more probable in sequences with a high proportion of serine, threonine and proline residues. Determining and analyzing the determinants of prevalence of O-glycosylation sites in the protein sequence is an essential step towards establishing the roles that these glycans play in health and disease. Statistical analysis and techniques like correlation and regression are performed on the 242 glycosylated protein sequences present in the database, O-GLYCBASE. The present paper deals with the analysis and results thereof.
 

 A Hybrid Optimization Approach
for Automated Parameter Estimation Problems
 

Carlos Quintero,¹ Miguel Argaez,¹ Hector Klie,¹

Leticia Velazquez,^1,2 and Mary Wheeler¹

(1) Department of Mathematical Sciences, The ¹ú²ú͵ÅÄ of Texas at El Paso

(2) Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas at El Paso

Abstract

We present a hybrid optimization approach for solving automated parameter estimation problems that is based on the coupling of the Simultaneous Perturbation Stochastic Approximation (SPSA) and a globalized Newton-Krylov Interior Point algorithm (NKIP) presented by Argáez et al. The procedure generates a surrogate model that yield to use efficiently first order information and applies NKIP algorithm to find an optimal solution. We implement the hybrid optimization algorithm on a simple test case, and present some preliminary numerical results.

 

 A User-friendly Database for Pseudoknots - RNAVBase-PK
()

Vindhya Shatdarsanam and Naresh Prodduturi

Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas at El Paso
 

Abstract

Among the RNA secondary structures, Pseudoknots are complex to understand and decipher. They have many biological functions, some already known and some still under research. To help these researchers by providing the information on these structures, we have developed a database which is similar to the database developed at Leiden ¹ú²ú͵ÅÄ, Pseudobase. But, this database RNAVBase-PK comes with many additional features which make it more user-friendly. This database contains all the information contained by Pseudobase with the additional features that allow users to search, select, format and visualize Pseudoknots. The database links each pseudoknot to the GenBank or EMBL record of the corresponding nucleotide sequence and can invoke PseudoViewer for automated graphical display of secondary structures. It also comes with a tool that helps for adding new pseudoknots in a user-friendly format. Our goal is to bring together information from various sources about a pseudoknot structure onto a single platform and make the information easily accessible to all its users. We shall further develop and expand this tool so that it will continue to fulfill the requirements of the growing research on RNA structures and functions.

Identification and Characterization of Small Molecules

That Specifically Inhibit FKBP52 Regulation of

Steroid Hormone Receptors
 

Dedeepya Vaka,^1,2 Heather Balsiger,² and Marc B. Cox²

(1) Bioinformatics Program, The ¹ú²ú͵ÅÄ of Texas at El Paso

(2) Border Biomedical Research Center and Department of Biological Sciences, The ¹ú²ú͵ÅÄ of Texas at El Paso
 

Abstract

Steroid hormone receptors require the ordered assembly of various chaperone and cochaperone proteins in order to reach a functional state. The final stage in the receptor maturation process requires the formation of a mutimeric complex consisting of an Hsp90 dimer, p23, and one of several large immunophilins (FKBP51, FKBP52, Cyp40, and PP5). All of the studies conducted by our laboratory and others suggest that, unlike other Hsp90-associated cochaperones, the immunophilins associate and/or regulate preferentially depending upon which client protein is present in the complex. For example, the large FK506-binding protein FKBP52 preferentially regulates androgen, progesterone, and glucocorticoid receptor-mediated signal transduction. Consistent with these findings, male FKBP52 knockout (52KO) mice display characteristics of partial androgen insensitivity and female 52KO mice have implantation defects related to progesterone receptor insensitivity. Thus, FKBP52 represents an attractive therapeutic target for the treatment of diseases that are dependent upon a functional hormone signaling pathway (e.g. prostate cancer). We developed a yeast-based screening assay for use in identifying small molecules that specifically inhibit FKBP52 regulation of androgen receptor function. We then used this assay to screen a diversified compound library containing approximately 2000 compounds of known structure that were selected to have representative diverse chemical structures. These screens resulted in the identification of two candidate compounds that potently and specifically inhibit FKBP52-mediated potentiation of receptor function in yeast. We are currently characterizing the inhibitory effects and specificity of inhibition of the candidate inhibitors in mammalian cells. Future studies, through the use of prediction modeling and various experimental approaches, will be aimed at identifying the inhibitor binding site(s) on FKBP52 and the active sites (pharmacophores) on the molecules.

 Expression of Cruzain: A Major Cysteine Protease in T. Cruzi
 

Nobish Varghese

Bioinformatics Program and Department of Biological Sciences

The ¹ú²ú͵ÅÄ of Texas at El Paso
 

Abstract

Infection by the protozoan parasite Trypanosoma Cruzi (T. Cruzi) results in Chagas disease which is the principal cause of early death out of heart disease in Latin America. Currently available treatment methods are generally unsatisfactory; the drugs used are highly toxic and ineffective, particularly in chronic stages of the disease. Thus there is an urgent need to develop novel, economic and effective drugs against the parasite. Our study focuses on the folding pathway of the cysteine protease cruzain, a key metabolic enzyme that plays a significant role in the growth and pathogenicity of T. Cruzi. Folding studies provide structural information necessary to develop novel classes of inhibitors against cruzain. Cruzain is currently being expressed and purified so that its folding pathway can be characterized.