141 research outputs found
MMsPred: a bioactivity and toxicology predictive system
No full textIn the last decade, the development and use of new methods in combinatorial chemistry and high-throughput screening has dramatically increased the number of known biologically active compounds. Paradoxically, the number of drugs reaching the market has not followed the same trend, often because many of the candidate drugs present poor qualities in absorption, distribution, metabolism, excretion, and toxicological properties (ADME-Tox). The ability to recognize and discard bad candidates early in the drug discovery steps would save lost investments in time and money. Machine learning techniques could provide solutions to this problem.
The goal of my research is to develop classifiers that accurately discriminate between active and inactive molecules for a specific target. To this end, I am comparing the effectiveness of the application of different machine learning techniques to this problem.	As a source of data we have selected a set of PubChem's public BioAssays1. In addition, with the objective of realizing a real-time query service with our predictors, we aim to keep the features describing the chemical compounds relatively simple.
At the end of this process, we should better understand how to build statistical models that are able to recognize molecules active in a specific bioassay, including how to select the most appropriate classification technique, and how to describe compounds in such a way that is not excessively resource-consuming to generate, yet contains sufficient information for the classification. We see immediate applications of such technology to recognize compounds with high-risk of toxicity, and also to suggest likely metabolic pathways that would process it
Effect of Non-Coding RNA on Post-Transcriptional Gene Silencing of Alzheimer Disease
No full textA large amount of hidden biological information is contained in the human genome, which is not expressed or revealed in the form of proteins; the usual end product form of gene expression. Instead, most of such information is in the form of non-coding RNAs (ncRNAs). ncRNAs correspond to genes that are transcribed, but do not get translated into proteins. This part of the genome was, till recently, considered as ‘junk’. The term ‘junk’ implied lack of any discernible function of these RNA. More than 98% of the human genomic size encompasses these non-coding RNAs. But, recent research has evidently brought out the indispensible contribution of non-coding RNA in controlling and regulating gene expression. ncRNA such as siRNAs and microRNAs have been reported to greatly help in causing post-transcriptional gene silencing (PTGS) in cells through RNA interference (RNAi) pathway. In this work, we have investigated the possibility of using siRNAs and microRNAs to aid in gene silencing of early onset Alzheimer’s disease genes. 
Alzheimer’s disease specific mutations and their corresponding positions in mRNA have been identified for six genes; Presenilin-1, Presenilin-2, APP (amyloid beta precursor protein), APBB3, BACE-1 and PSENEN. 

Small interfering RNAs (siRNAs) that can cause PTGS through RNA interference pathway have been designed. RNA analysis has been done to verify complementarity of antisense siRNA sequence with target mRNA sequence. Interaction studies have been done computationally between these antisense siRNA strands and seven Argonaute proteins. From the interaction studies, only one of the seven Argonaute proteins; 1Q8K, was found to have interaction with the siRNAs indicating the importance and uniqueness of this particular protein in RISC (RNA induced silencing complex). 

The interaction studies have been carried out for the microRNAs also. Out of the 700 mature human microRNAs collected, 394 microRNAs have been identified to show partial complementarity with their target sequence on PSEN-1 mRNA. Of these 394, five microRNAs have shown partial complementarity to early onset Alzheimer’s disease specific mutations in PSEN-1 mRNA. Interaction studies have been done between these microRNAs and Argonaute proteins. Thus, design, characterization and analysis of ncRNAs that contribute to post transcriptional gene silencing of Alzheimer’s disease have been achieved.

CoryneRegNet: An ontology-based data warehouse of corynebacterial transcription factors and regulatory networks
No full textBaumbach J, Brinkrolf K, Czaja LF, Rahmann S, Tauch A. CoryneRegNet: An ontology-based data warehouse of corynebacterial transcription factors and regulatory networks. BMC Genomics. 2006;7(1): 24
- …
