ToolsAbout

Colour an alignment

Colours an aligned FASTA by conservation or by amino acid property, with a conservation bar.

Coloured alignment
Two ways to colour an alignment
Conservation colours every column by how much the sequences agree there, so the eye goes straight to the blocks that have not changed: active site residues, the core of a fold, a binding motif. Properties colours each residue by its side chain instead, which shows whether a column that is not conserved is at least chemically consistent, a hydrophobic position that tolerates any of I, L, V or M.
Under the alignment sit two summaries of the same columns: a bar whose height is the conservation of that column, and a symbol line in the Clustal convention.
How conservation is calculated
conservation % = sequences carrying the most common residue ÷ sequences × 100
Gaps are in the denominator. A column of ten sequences in which eight carry K and two are gaps is 80 % conserved, not 100 %, which is what you want when a ragged end or one long insertion leaves a column nearly empty. It is the same convention theidentity and similarity calculator and thealignment trimmer use, so the numbers on all three pages agree. For the information content of a column in bits, which weighs rare residues differently, use the Shannon entropy calculator.
SymbolMeaning
*One residue in every sequence, no gaps
:All residues from one strongly similar group, no gaps
.All residues from one weakly similar group, no gaps
The strong groups are STA, NEQK, NHQK, NDEQ, QHRK, MILV, MILF, HY and FYW; the weak groups are CSA, ATV, SAG, STNK, STPA, SGND, SNDEQK, NDEQHK, NEQHRK, FVLIM and HFY. These are the sets Clustal uses, derived from the Gonnet matrix.
The property groups
GroupResidues
HydrophobicA V L I M F W
Polar unchargedS T N Q C Y
AcidicD E
BasicK R H
Glycine and prolineG P
Glycine and proline get their own colour because they are the two residues whose effect is structural rather than chemical: glycine has no side chain and gives the backbone freedom, proline locks it. A nucleotide alignment is coloured by base instead, A green, C blue, G orange and T or U red.
Worked example
The example is a family of five aligned protein sequences, 38 columns wide, around a P-loop motif. 22 of the 38 columns are fully conserved, marked with an asterisk, and mean conservation is 87.9 %. The interesting column is 27: one sequence has an A there and the other four have a gap, so it scores 20 %, and the bar drops almost to nothing. A tool that ignored gaps would call that column perfectly conserved.
Reading a coloured alignment
  • A solid block of colour across all sequences is a conserved motif. Check it against the literature before deciding it is an active site.
  • One sequence breaking a conserved column is often a sequencing or annotation error, not biology.
  • A column that changes colour in Properties but keeps a high conservation score is a conservative substitution; a column that keeps one colour while the letters change is a position under chemical rather than exact constraint.
  • Long stretches of gaps mean the alignment, not the protein, needs attention. Trim them with the alignment trimmer.
Frequently asked questions
How is conservation measured for each column?
It is the share of sequences that carry the most common residue in that column, gaps included in the denominator. A column with eight identical residues and two gaps is 80 percent conserved, not 100 percent. Counting it the other way, over the residues present only, makes a column with one residue and nine gaps look perfectly conserved, which is the usual reason two tools disagree about a gappy alignment.
What do the symbols under the alignment mean?
They follow the Clustal convention. An asterisk marks a column where every sequence has the same residue and none has a gap. A colon marks a column with no gaps whose residues all come from one strongly similar group, such as I, L, V and M. A full stop marks one weakly similar group. A blank means neither, or that the column contains a gap.
Which amino acids count as hydrophobic, polar, acidic or basic?
Hydrophobic: A, V, L, I, M, F and W. Polar uncharged: S, T, N, Q, C and Y. Acidic: D and E. Basic: K, R and H. Glycine and proline are coloured separately because their behaviour is set by their backbone rather than by their side chain. Cysteine is grouped with the polar residues, as in most textbooks, although in a folded protein it often sits in a hydrophobic environment.
Does it work on a DNA alignment?
Yes. Conservation is measured the same way for nucleotides. On the Properties tab a nucleotide alignment is coloured by base instead of by side chain, with the usual green, blue, orange and red for A, C, G and T.
Do the sequences have to be aligned?
Yes. Colouring by column only means something when the columns mean something, so the input must be an aligned FASTA in which every record has the same length, gaps included. Records of other lengths are padded with gaps on the right so that something is still drawn, and a warning says which ones.