From the Origins of SARS-CoV-2 to the Correct Application of BLAST / 中国生物化学与分子生物学报

Basic local alignment search tool (BLAST) is one of the popular sequence similarity analysis tools. However, some students and researchers just blindly use the default parameters. Moreover, some students are confused about how to choose the right program. In a word, it is prone to be misused and researchers often draw conclusions incorrectly. In view of this, we traced back the internet hot topic in early 2020 - "MORDERATELY STRONG CONFIRMATION OF A LABORATORY ORIGIN OF COVID-19", and took it as teaching materials to guide the student to use BLAST currently through reanalyzing and reproducing the source of errors. Then we arranged an interesting experiment about fabricating dinosaur genes through modifying a chicken gene. In the experimental design to make the students grasp the BLAST tools better, one group fabricated the dinosaur gene and the other group decrypted the added bases. This instructional design could be conducive to cultivate students ' ability about distinguishing different viewpoints correctly, and we hope it can be enlightening and helpful to the teaching of BLAST tools.

Prediction of protein subcellular locations by ensemble of improved K-nearest neighbor / 生物工程学报

Adaboost algorithm with improved K-nearest neighbor classifiers is proposed to predict protein subcellular locations. Improved K-nearest neighbor classifier uses three sequence feature vectors including amino acid composition, dipeptide and pseudo amino acid composition of protein sequence. K-nearest neighbor uses Blast in classification stage. The overall success rates by the jackknife test on two data sets of CH317 and Gram1253 are 92.4% and 93.1%. Adaboost algorithm with the novel K-nearest neighbor improved by Blast is an effective method for predicting subcellular locations of proteins.

A Heuristic Algorithm to Find All Normalized Local Alignments Above Threshold

Local alignment is an important task in molecular biology to see if two sequences contain regions that are similar. The most popular approach to local alignment is the use of dynamic programming due to Smith and Waterman, but the alignment reported by the Smith-Waterman algorithm has some undesirable properties. The recent approach to fix these problems is to use the notion of normalized scores for local alignments by Arslan, Egecioglu and Pevzner. In this paper we consider the problem of finding all local alignments whose normalized scores are above a given threshold, and present a fast heuristic algorithm. Our algorithm is 180-330 times faster than Arslan et al.''s for sequences of length about 120 kbp and about 40-50 times faster for sequences of length about 30 kbp.