Multipe Sequence Alignments:
Probing of the Twilight Zone

  1. Fetching the sequences

  2. Extracting Homologous Segments

  3. Making an Alignment

    • Cut and paste the sequence in the multiple alignment server of your choice
    • Here is an alignment of the two segments: [two]
    • Here is an alignment of the three segments: [three]
    • The difference between the two alignments are obvious:[comparison]

  4. Testing the validity of the model

    • Add extra sequences to check if they affect the original alignment
    • Extra sequences will be found using blast
    • Sequences must be as diverse as possible, with key residues well conserved
    • Here is an example of a correct alignment [ref_aln]
    • Observe that also the sequences are distantly related they remain conserved on key positions

    Although the similarity between hmga_chite.fasta and hmgl_trybr.fasta is clearly within the twilight zone, the multiple alignment shows that they share key positions of the hmg domain. As a consequence it is quite likely that these parts of the proteins are indeed homologous.

Questions should be sent to C.Notredame