For genome-wide contigs, misassemblies and translocations almost always happen, which will break glocal alignment; if contigs can really be aligned in one piece, proper local alignment can identify the correct alignment at a high chance. There is no need to try glocal alignment. As to programs, MUMmer is usually the first choice for bacterial (btw, MUMmer reports local hits). Cross_match also works well for me in small scale. With FASTA, you cannot align >40kb query, IIRC. BLAST is slow and usually produces very fragmented alignments under the default setting which are hard to process. I would not use these two. I have heard good things about Mauve, but have not tried myself. There are also bwa-sw and bwa-mem for mammalian genomes (my programs), but they are less used.
i have a fasta file with 113 contigs and a reference sequence in fasta file. i have to align my contigs to the reference sequence and i have to see in which order my contigs are aligned to the reference sequence. pls can you help me which software i can use for this ?
2 answers
You have to make up your mind: you can either use global-local alignment or local alignments. If you are quite sure, the contigs are not misassembled, you can use global-local alignment. You have only few contigs, therefore you can possibly use glsearch or if that doesn't work MUMer. On the other hand you won't necessarily find misassembled contigs that way, e.g. a contig that alignes only partially in two or more sections, and without testing for such contigs, you cannot know if your assembly is good. Therefore, use also local alignment tools, e.g. FASTA or BLAST.
Log in to answer this question.