Hi, Here is an abstract from Soumya Sir about the talk on data mining to be presented at the forthcoming ILUG-Delhi meet this Sun., 1st April. It looks interesting, especially as it has applications in other areas than just bio-informatics.
Regards, Gora -------- Forwarded Message -------- From: Soumyadeep nandi <[EMAIL PROTECTED]> To: [EMAIL PROTECTED] Cc: Dr. Andrew Michael Lynn <[EMAIL PROTECTED]> Subject: Abstract: Data mining with Open Source tools: Application to Biological Sequence Classification. Date: Wed, 28 Mar 2007 07:37:11 -0700 (PDT) Data mining with Open Source tools: Application to Biological Sequence Classification. Computational prediction of protein function or annotation as a whole is a challenging problem. The assignment of protein function is performed by comparing a sequence with unknown function to sequences of known function. There are number of methods applied in present days annotation techniques like sequence similarity search (viz FASTA, BLAST), profile based search (PSI-BLAST, HMMER) etc. The profile based techniques using hidden markov model proved to be superior to sequence similarity based methods in detecting remote homologs. Other methods used for general discrimination are NaiveBayes, Support Vector Machine, Artificial Neural Network, Decision tree. Recent applications extend the use of the support vector machines to protein homology detection. This technique performed significantly better in image recognition and speech recognition. In this demo we demonstrate the application of data mining with the open source machine learning software WEKA to sequence classification. Regards, Soumyadeep ________________________________________________________________________ The fish are biting. Get more visitors on your site using Yahoo! Search Marketing. _______________________________________________ ilugd mailinglist -- ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/ilugd@lists.linux-delhi.org/