DNA Sequence Classification in the Presence of Sequencing Errors Project of CSE847

We propose a new method to solve the problem of DNA sequence classification in the presence of sequence errors. This method incorporates sequencing error models into standard Viterbi Algorithm(AJ, 1967) which is used to find the best path that a query DNA sequence is generated by a profile Hidden Markov Model (profile HMM)(SR., 1998). Our method will correct the sequencing errors and then align the corrected query sequence to different protein families. By sequencing error correction our method will improve the prediction accuracy for sequence classification. Moreover, we will classify more sequences to one of the candidate families which cannot be classified by current tools due to sequencing errors.