Dali Tutorial

1. Introduction
2. Inputs
2.1. Submission
3. Outputs
3.1. Structural alignment view
3.1.1.Stacked sequence logos
3.2. 3D superimposition view
3.3. Integrated sequence search tools
3.4. Visualization of protein structure space
4. Example
4.1. Selecting the input set
4.2. Position of the amidohydrolase and PHP superfamilies in structure space
4.3. Overlapping sequence motifs of the amidohydrolase and PHP superfamilies
4.4. Looking into the PHP superfamily
4.5. Diversity of molecular functions in the amidohydrolase superfamily
4.6. Conclusion
5. Downloads

Appendix A: Sample PDB entry

Appendix B: Input data for plot in Figure 10

1 Introduction
Dali is a protein structure comparison server. The server has been running continuously for over 20 years.
The server operated first in Heidelberg (Germany), then Hinxton (UK), now Helsinki (Finland). Dali is based
on distance matrix comparison (see References for methods). In favourable cases, structure comparison
can reveal distant evolutionary relationships not seen by sequence comparison.

The server (http://ekhidna2.biocenter.helsinki.fi/dali/) supports three types of requests:

1. PDB search
2. Pairwise comparison
3. All against all comparison

The server takes the 3D coordinates of protein structures as input and returns a list of similar structures,
structural alignments and superimposed structures. The all against all comparison also returns a structural
dendrogram and a projection from protein structure space. The results are linked to sequence search and
function prediction servers.

This tutorial explains the web interface of the Dali server using live examples.

2 Inputs
The PDB format is based on records with keywords. A sample PDB structure is given in Appendix A. Only
ATOM records are required by Dali. The full specification of the format can be found at
http://www.wwpdb.org/docs.html.

The following restrictions apply:

1
  The structure must contain the coordinates of the backbone atoms: N, CA, C and O. If your
structure has only the C-alpha coordinates, you can generate a complete backbone using the
MaxSprout server at http://www.ebi.ac.uk/maxsprout.
 The structure must contain at least 30 residues. Shorter chains are ignored by Dali.

Publicly available repositories of protein structures are RCSB, PDBe, and PDBj.

PDB entries have a PDB identifier, which is four characters long and consists of a digit followed by three
letters or digits, for example, 3ubp. You can find the PDB entries matching a keyword search at RCSB,
http://www.rcsb.org/. Each entry can contain one or more chains. The chain identifier is one character. For
example, PDB entry 3ubp has three chains A, B and C. In the submission forms, the chain identifier must be
concatenated with the PDB identifier, for example, 3ubpC. The PDB search submission form gives hints on
possible continuations when you start typing the PDB identifier.

The Dali server does not accept an amino acid sequence as input. If you know only the amino acid sequence
of your protein, you can search for a related PDB structure using sequence comparison with servers like
SANSparallel (http://ekhidna2.biocenter.helsinki.fi/sans/, search against PDB database). Comparative
modeling servers like SwissModel generate a model which only replaces the side chains (according to a
sequence alignment) while the backbone stays very close to the template structure. More adventurous
servers may generate a model ab initio when the query sequence has no obvious homolog of known
structure. For example, PHYRE, I-TASSER and ROBETTA have been some of the top performers in CASP
(Critical Assessment of Structure Prediction).

2.1 Submission
The submission forms (Figure 1) for PDB search, pairwise comparison and all against all comparison accept
one, two to 11, and three to 64 input structures, respectively. In pairwise and all against comparison, you
must click on the +/- buttons to create the required number of input fields. All against all comparison has
two alternative submission forms, one for input sets with uploaded structures and an alternative one for
input sets composed only of PDB identifiers.

PDB searches and all against all comparisons are time consuming, upwards of 10-15 minutes. A queueing
system is in use, so you have to wait even longer if there are many simultaneous requests. If you left your
email address, you will receive an email notification when the results are ready. Otherwise you must stay
on the result page or bookmark it so that you can return to it later.

Figure 1. Submission forms.

2
3 Outputs
In this tutorial, we use the amidohydrolase superfamily as example. The amidohydrolase superfamily was
first discovered based on structural similarities between urease, adenosine deaminase and
phosphotriesterase. Let’s see how they are related structurally.

1. Go to the submission form for pairwise comparison (Figure 1 middle; from the main page click on
the “Pairwise” tab).
2. If you do not know the PDB identifiers of the structures you are interested in, make a detour to
RCSB’s keyword search. For example, from PDB Text we find a phosphotriesterase entry 4xd3 with
chains A and G.
3. Type 4xd3A in the box for first protein structure.
4. Press the plus button twice to create two input fields for second structures.
5. Type 3ubpC and 1a4mA in the boxes for second protein structures.
6. Press submit.
7. Wait for the result.

All request types produce match lists in the same format (Figure 2). The matches are sorted by Dali Z-score.
The number column has hyperlinks to the pairwise structural alignment between the query and match
structure. Each matched structure is also hyperlinked to the PDB entry. The coordinates of the PDB entry
are superimposed on the query structure by rigid-body rotation and translation. The checkboxes are used
to select a subset of matches for interactive visualization. The buttons above the match list launch
structural alignments in 1D (“Structural Alignment”) or 3D (“3D Superimposition (PV)”), and provide links to
sequence analysis tools (“SANS” sequence database search, “PANZ” function prediction). The following
sections demonstrate the interactive visualization options. Use your result from the above exercise or this
live example.

Figure 2. Result from pairwise comparison.

3.1 Structural alignment view

The multiple alignment view opens in a new window and displays the alignment of the query structure and
the selected matches. The upper block shows the amino acid sequences and the lower block the secondary
structure states (H: helix, E: sheet, L: coil). The most frequent symbol in each column is coloured. The
alignment view has an option ‘expand gaps’. If the option is checked, the complete sequence of all proteins
is shown. Residues without a match in the query structure are shown in lowercase. If the option is not
checked, all matching sequences are shown stacked on the query sequence, and insertions relative to the
query sequence are hidden (Figure 3). We are comparing distantly related proteins but they have a striking
signature of invariant amino acids, including the histidines at position 20 and 22 in Figure 3.

3
Figure 3. Structural Alignment view (Expand gaps option off).

3.1.1 Stacked sequence logos

The structural alignment view only considers the amino acid sequences of the selected protein structures.
The button labelled “Show Stacked Sequence Logos” generates a sequence profile around each structurally
known protein by searching for homologs in Uniprot. The sequence profiles are then showed stacked
against the query protein (Figure 4, live example). The sequence profiles require optimizing an HMMer
model for each structure and this is a slow process, taking about 8 seconds per logo. When you are
displaying many logos, zoom out with CTRL-minus so that more rows fit on the screen. Zoom back in with
CTRL-plus. The stacked sequence logos show which positions are conserved across the whole set rather
than matching by chance in the structural alignment of a small number of structures. The sharp sequence
signature motifs often pinpoint the active site of enzyme superfamilies (Figure 5).

Figure 4. Showing a section of Stacked Sequence Logos view (Expand gaps option on).

3.2 3D superimposition view

The 3D superimposition view opens in a new window and displays the selected structures superimposed on
the query structure (Figure 5). This is a rigid body transformation, which may look ugly if the RMSD is high.
(Dali does not optimize RMSD, it matches contacts). The web page has simple toggles to select display

4
styles. In particular, sequence and structure conservation can be mapped to the query structure and shown
by colour. The colour map for conservation mapping goes from blue for the highest values through green to
red for the lowest values. Sequence conservation is calculated as the relative entropy of a column, SUM p(i)
log(p(i)/q(i)), where the sum is over twenty amino acid types I and p(i)=n(i)/N where n(i) is the number of
occurrences and N is the number of rows in the alignment, and q(i) are the frequencies of amino acid types
in the sequence database. The logarithm is taken in base 2 so the unit of relative entropy is bits. Structure
conservation is simply the fraction of selected structures that are structurally aligned to the query
structure.

We use PV as structure viewer. PV is a Javascript based viewer which works on modern browsers. PV works
as advertised with Chrome and Firefox. With Internet Explorer there are some quirks with asynchronous
refreshing of the image, which tends to disappear altogether after the user clicks options but can be
restored by moving the cursor or clicking the show/hide options repeatedly.

1. Check 3ubpC and 1a4mA in your summary page and press the “3D Superimposition (PV)” button.
Alternatively, you may use this live example.
2. Scale the window to full screen size. This places the viewer area (with light blue passepartout) and
option checkboxes side by side.
3. You should see a spaghetti of multiple C-alpha traces and side chains. Click on the radio button labelled
“Cartoon”. Uncheck “All” and check “Query” in the Show/hide structures options and you should see a
green cartoon representation of the query protein.
4. Click on “Structure conservation” for Query colour. Dark blue regions are structurally aligned in all
three structures. Hold down the left mouse button and move the cursor in the viewer area to rotate
the structure. Hold down the middle button and move the cursor up/down to zoom in/out.
5. Switch to “Sequence conservation” for Query colour and check “Query” in Show/hide side chains
options.
6. Clear up the messy picture by removing less conserved side chains. Move sliding ruler under “Query
side chains > 0 bits” to the left until the value is 4.15 bits. Now conserved residues at the active site are
highlighted. Click on the side chains to see their labels with residue numbers.
7. Check the Show/hide ligands option. Click atoms to see their labels.

Figure 5. Screenshot of protein viewer (PV).

5
Figure 6. The results of all against all comparison are divided under five tabs.

6
3.3 Integrated sequence search tools
Sometimes there are uncharacterized proteins in the summary list. From the interactive summary (Figure 2)
you can send the amino acid sequences of selected subsets to search Uniprot by SANSparallel (SANS
button) or predict function by PANNZER2 (PANZ button). These tools do not use structure information in
any way and they are provided for convenience. The mapping of structures to Uniprot brings the great
advantage of crosslinks to literature and protein family classification resources (e.g. PFAM).

3.4 Visualization of protein structure space

All against all comparison generates an overview of protein structure space for a set of input structures.
The Example section below will take you through the discovery process for structural classification with
Dali. At this point, we just show the outputs. The output of all against all comparison has a different layout
from PDB search and pairwise comparison. In addition to the summary lists, there are tabs for plots
generated from the all against all similarity matrix (Figure 6 and live example).

The dendrogram is clickable. The leaves are linked to the summary list of that structure, which shows the
structural alignments of the other structures of the input set aligned against it. The similarity matrix and
scatterplot are interactive, responding to hovering the mouse over a data point. A toolbar appears at the
upper right corner of the scatterplot. Click on the single arrow to see the label attached to the nearest data
point. Structural alignment summaries are linked both to the leaves of the dendrogram and the list under
the Summaries tab. The similarity matrix, eigenvector analysis results and pseudo-phylogenetic trees in
Newick and PhyloXML format are available under the Download tab. Appendix B shows an example where
the eigenvectors from correspondence analysis were downloaded and grouped according to branches in
the dendrogram for plotting in Excel.

4 Example
In this example, we revisit the amidohydrolase superfamily 20 years after it was first discovered. We do
PDB searches to find a representative set of current members. We shed light on the relationship between
amidohydrolases and the PHP superfamily, which has intriguing structural and sequence similarities to
amidohydrolases. We do all against all comparison to get an overview of the position of these two
superfamilies in protein structure space. The final result can be seen in this live example.

4.1 Selecting the input set

We already know about three structures in the amidohydrolase superfamily from section 3. Using these
structures as queries in PDB searches, we can collect more members from the current PDB. To decide
whether a protein is a member or not, we look at characteristic features of the superfamily such as: are the
nearest structural neighbours all amidohydrolases? is the structural core conserved? is the sequence
signature conserved? Dali’s search algorithm uses heuristics and is not guaranteed to deliver the optimal
alignment. Therefore we performed a number of searches from diverse seeds and combined the results.

Figure 7 shows the result from one PDB search. All the structures listed in Figure 7 are legitimate members
of the amidohydrolase superfamily, despite having diverse molecular functions. PDB search results are
reported for PDB90 and PDB100. PDB100 contains all structures in the PDB. PDB90 is a non-redundant
subset of PDB structures with less than 90 % sequence identity to each other. PDB and PDB90 are updated
weekly. PDB90 representatives are not necessarily stable from week to week.

Merging the structural neighbour lists of a few seed structures and removal of redundant structures (with
more than 30-40 % sequence identity) resulted in the set shown in Figure 8. The set includes members of
the amidohydrolase superfamily and of the PHP superfamily. The PHP superfamily has structural similarities
to amidohydrolases and partially overlapping sequence motifs. Tryptophan synthase was added as an
outgroup which is not thought to be related to either of the aforementioned superfamilies.

7
 # Query: 4xd3A
# No: Chain Z rmsd lali nres %id PDB Description
1: 1qw7-A 59.4 0.5 327 336 96 MOLECULE: PARATHION HYDROLASE;
2: 3a3w-A 58.3 0.7 326 329 87 MOLECULE: PHOSPHOTRIESTERASE;
3: 4if2-A 43.2 1.7 308 324 36 MOLECULE: PHOSPHOTRIESTERASE HOMOLOGY PROTEIN;
4: 2vc5-A 42.6 1.8 305 314 34 MOLECULE: ARYLDIALKYLPHOSPHATASE;
5: 4rdz-B 42.1 1.8 305 316 30 MOLECULE: PARATHION HYDROLASE;
6: 2zc1-A 41.3 2.1 303 333 33 MOLECULE: PHOSPHOTRIESTERASE;
7: 5ch9-A 41.1 2.2 307 328 31 MOLECULE: PHOSPHOTRIESTERASE;
8: 3k2g-B 39.2 2.2 295 358 29 MOLECULE: RESINIFERATOXIN-BINDING, PHOSPHOTRIESTERASE-
9: 3rhg-A 38.9 2.2 296 363 27 MOLECULE: PUTATIVE PHOPHOTRIESTERASE;
10: 3pnz-A 38.2 2.5 294 329 28 MOLECULE: PHOSPHOTRIESTERASE FAMILY PROTEIN;
11: 3msr-A 37.1 2.1 288 353 25 MOLECULE: AMIDOHYDROLASES;
12: 1bf6-A 36.7 2.1 276 291 30 MOLECULE: PHOSPHOTRIESTERASE HOMOLOGY PROTEIN;
13: 3guw-A 19.8 2.5 213 233 17 MOLECULE: UNCHARACTERIZED PROTEIN AF_1765;
14: 1j6o-A 19.8 3.1 237 260 14 MOLECULE: TATD-RELATED DEOXYRIBONUCLEASE;
15: 1zzm-A 19.8 3.4 236 259 18 MOLECULE: PUTATIVE DEOXYRIBONUCLEASE YJJV;
16: 4p5u-A 19.8 3.0 232 262 15 MOLECULE: TAT-LINKED QUALITY CONTROL PROTEIN TATD;
17: 3rcm-A 19.7 3.1 236 279 16 MOLECULE: TATD FAMILY HYDROLASE;
18: 1yix-A 19.6 3.2 234 265 18 MOLECULE: DEOXYRIBONUCLEASE YCFH;
19: 2gzx-A 19.4 3.2 232 253 16 MOLECULE: PUTATIVE TATD RELATED DNASE;
20: 3ipw-A 19.3 3.3 243 301 12 MOLECULE: HYDROLASE TATD FAMILY PROTEIN;
21: 2y1h-B 19.2 3.2 230 265 16 MOLECULE: PUTATIVE DEOXYRIBONUCLEASE TATDN3;
22: 1onx-A 19.1 3.0 235 390 17 MOLECULE: ISOASPARTYL DIPEPTIDASE;
23: 2xio-A 19.0 3.1 238 293 11 MOLECULE: PUTATIVE DEOXYRIBONUCLEASE TATDN1;
24: 3be7-A 18.7 3.6 243 399 15 MOLECULE: ZN-DEPENDENT ARGININE CARBOXYPEPTIDASE;
25: 3mkv-A 18.7 3.7 246 414 15 MOLECULE: PUTATIVE AMIDOHYDROLASE;
26: 2qs8-A 18.6 3.7 252 407 17 MOLECULE: XAA-PRO DIPEPTIDASE;
27: 3mtw-A 18.1 3.8 244 404 14 MOLECULE: L-ARGININE CARBOXYPEPTIDASE CC2672;
28: 2ftw-A 18.0 3.1 245 484 13 MOLECULE: DIHYDROPYRIMIDINE AMIDOHYDROLASE;
29: 2vr2-A 18.0 2.9 240 478 13 MOLECULE: DIHYDROPYRIMIDINASE;
30: 2vm8-A 18.0 3.2 246 477 10 MOLECULE: DIHYDROPYRIMIDINASE-RELATED PROTEIN 2;
31: 3e2v-A 17.9 3.8 244 363 10 MOLECULE: 3'-5'-EXONUCLEASE;
32: 4b3z-D 17.9 3.4 250 477 12 MOLECULE: DIHYDROPYRIMIDINASE-RELATED PROTEIN 1;
33: 4cnu-A 17.9 3.4 250 488 12 MOLECULE: DIHYDROPYRIMIDINASE-LIKE 3;
34: 4gz7-A 17.8 3.3 247 492 11 MOLECULE: DIHYDROPYRIMIDINASE;
35: 3dc8-A 17.6 3.3 248 483 15 MOLECULE: DIHYDROPYRIMIDINASE;
36: 4l9x-B 17.6 3.9 243 458 17 MOLECULE: TRIAZINE HYDROLASE;
37: 1gkr-A 17.5 3.6 240 451 13 MOLECULE: NON-ATP DEPENDENT L-SELECTIVE HYDANTOINASE;
38: 3giq-A 17.5 3.1 243 475 17 MOLECULE: N-ACYL-D-GLUTAMATE DEACYLASE;
39: 4b90-A 17.5 3.2 242 485 14 MOLECULE: DIHYDROPYRIMIDINASE-RELATED PROTEIN 5;
40: 4c5y-A 17.5 3.6 237 436 15 MOLECULE: OCHRATOXINASE;
41: 1nfg-A 17.4 3.4 245 457 16 MOLECULE: D-HYDANTOINASE;
42: 3cjp-A 17.3 2.6 212 262 13 MOLECULE: PREDICTED AMIDOHYDROLASE, DIHYDROOROTASE FAMILY;
43: 1m7j-A 17.2 3.2 245 474 16 MOLECULE: D-AMINOACYLASE;
44: 2qpx-A 17.2 3.1 235 376 10 MOLECULE: PREDICTED METAL-DEPENDENT HYDROLASE OF THE TIM-BA
45: 2p9b-A 17.1 3.8 239 407 16 MOLECULE: POSSIBLE PROLIDASE;
46: 2paj-A 17.1 3.4 230 421 14 MOLECULE: PUTATIVE CYTOSINE/GUANINE DEAMINASE;
47: 4i6k-A 17.1 3.4 233 267 11 MOLECULE: AMIDOHYDROLASE FAMILY PROTEIN;
48: 4tqt-D 17.0 3.4 249 481 13 MOLECULE: D-HYDANTOINASE;
49: 3gg7-A 16.8 3.4 220 243 18 MOLECULE: UNCHARACTERIZED METALLOPROTEIN;
50: 3d6n-A 16.8 3.7 245 422 17 MOLECULE: DIHYDROOROTASE;
51: 2ffi-A 16.7 3.4 236 273 13 MOLECULE: 2-PYRONE-4,6-DICARBOXYLIC ACID HYDROLASE, PUTATIV
52: 3s2j-A 16.5 3.4 241 393 12 MOLECULE: DIPEPTIDASE;
53: 2z00-A 16.4 3.5 239 426 16 MOLECULE: DIHYDROOROTASE;
54: 2vun-A 16.4 3.1 218 385 14 MOLECULE: ENAMIDASE;
55: 2oof-A 16.4 3.9 234 403 19 MOLECULE: 4-IMIDAZOLONE-5-PROPANOATE AMIDOHYDROLASE;
56: 2g3f-A 16.4 4.0 236 414 19 MOLECULE: IMIDAZOLONEPROPIONASE;
57: 3irs-A 16.4 3.5 226 281 11 MOLECULE: UNCHARACTERIZED PROTEIN BB4693;
58: 2i5g-A 16.2 3.3 236 325 13 MOLECULE: AMIDOHYDROLASE;
59: 2gok-A 16.2 4.1 233 404 19 MOLECULE: IMIDAZOLONEPROPIONASE;
60: 3b40-A 16.1 3.3 242 400 10 MOLECULE: PROBABLE DIPEPTIDASE;
61: 3lu2-A 16.1 3.2 229 309 12 MOLECULE: LMO2462 PROTEIN;
62: 4v1x-E 16.1 3.4 233 474 15 MOLECULE: ATRAZINE CHLOROHYDROLASE;
63: 2fty-A 15.8 3.5 245 532 13 MOLECULE: DIHYDROPYRIMIDINASE;
64: 3hm7-B 15.7 3.0 219 437 13 MOLECULE: ALLANTOINASE;
65: 3gri-B 15.6 3.4 231 423 11 MOLECULE: DIHYDROOROTASE;
66: 2z26-A 15.6 3.7 242 344 11 MOLECULE: DIHYDROOROTASE;
67: 3nqb-A 15.6 3.1 214 587 18 MOLECULE: ADENINE DEAMINASE 2;
68: 1itq-A 15.6 3.4 235 369 9 MOLECULE: RENAL DIPEPTIDASE;
69: 4lfy-B 15.6 3.7 241 353 11 MOLECULE: DIHYDROOROTASE;
70: 3e74-B 15.5 2.9 220 433 13 MOLECULE: ALLANTOINASE;
71: 3ly0-B 15.4 3.3 237 352 11 MOLECULE: DIPEPTIDASE AC. METALLO PEPTIDASE. MEROPS FAMILY
72: 2amx-A 15.4 3.5 233 364 12 MOLECULE: ADENOSINE DEAMINASE;
73: 2wjd-A 15.2 3.2 206 244 13 MOLECULE: TYROSINE-PROTEIN PHOSPHATASE CPSB;
74: 3lnp-A 15.2 3.9 229 441 16 MOLECULE: AMIDOHYDROLASE FAMILY PROTEIN OLEI01672_1_465;
75: 2dvt-A 15.0 3.5 221 325 16 MOLECULE: THERMOPHILIC REVERSIBLE GAMMA-RESORCYLATE DECARBO
76: 2ics-A 14.9 3.4 220 368 12 MOLECULE: ADENINE DEAMINASE;
77: 3ewd-A 14.9 4.1 235 364 11 MOLECULE: ADENOSINE DEAMINASE;
78: 2gwg-A 14.9 3.2 223 329 10 MOLECULE: 4-OXALOMESACONATE HYDRATASE;
79: 4dyk-A 14.9 3.7 227 437 12 MOLECULE: AMIDOHYDROLASE;
80: 1j6p-A 14.8 3.9 229 407 12 MOLECULE: METAL-DEPENDENT HYDROLASE OF
81: 3rys-A 14.8 4.1 233 335 13 MOLECULE: ADENOSINE DEAMINASE 1;
82: 3pnu-A 14.7 3.7 235 338 10 MOLECULE: DIHYDROOROTASE;
83: 1ie7-C 14.7 3.2 216 570 17 MOLECULE: UREASE GAMMA SUBUNIT;
84: 1a5k-C 14.7 3.2 217 566 17 MOLECULE: UREASE (GAMMA SUBUNIT);
85: 4icm-A 14.7 3.6 221 335 14 MOLECULE: 5-CARBOXYVANILLATE DECARBOXYLASE;

Figure 7. Top part of PDB90 summary list for PDB search.

8
 1a4mA adenosine deaminase
1a5kC urease gamma subunit
1bf6A phosphotriesterase
1bksA tryptophan synthase
1gkpA hydantoinase
1itqA renal dipeptidase
1j5sA uronate isomerase
1j6pA uncharacterized
1k6wA cytosine deaminase
1m65A probable phosphatase YcdX
1onxA isoaspartyl dipeptidase
1v77A ribonuclease P protein component 3
1yrrB N-acetylglucosamine-6-phosphate deacetylase
2a3lA AMP deaminase
2anuA predicted phosphoesterase
2dvtA gamma-resocylate decarboxylase
2ffiA 2-pyrone-4-6-dicarboxylic acid hydrolase
2gwgA 4-oxalomesaconate hydratase
2imrA S-adenosylhomocysteine deaminase
2ob3A parathion hydrlolase
2ogjA deacetylase
2oofA 4-imidazolone-5-propanoate amidohydrolase
2pajA isoxanthopterin deaminase
2qpxA uncharacterized
2uz9A guanine deaminase
2vc5A aryldialkylphosphatase
2vunA enamidase
2y1hB deoxyribonuclease TatD
2yb1A uncharacterized
3au2A DNA polymerase X
3cjpA uncharacterized
3dcpA histidinol-phosphatase
3e38A uncharacterized
3e74A allantoinase
3f2bA DNA polymerase polC
3gg7A uncharacterized
3giqA N-acyl-D-glutamate deacylase
3griA dihydroorotase
3iacA glucuronate isomerase
3icjA AepA exoenzymes regulatory protein
3irsA uncharacterized
3k2gB resiniferatoxin-binding phosphotriesterase
3ls9A triazine hydrolase
3mkvA uncharacterized
3mtwA L-arginine carboxypeptidase
3nqbA adenine deaminase
3ooqA uncharacterized
3pnuA dihydroorotase
3qy6A tyrosine-protein phosphatase
4b3zD dihydropyriminidase-related
4c5yA ochratoxinase
4cqbA N-isopropylammelide isopropyl amidohydrolase
4dlfA uncharacterized
4dziC uncharacterized
4hk5D uracil-5-carboxylate decarboxylase
4mupB D-galactarolactone isomerase
4ofcA 2-amino-3-carboxymuconate-6-semialdehyde decarboxylase
4qrnA 5-carboxyvanillate decarboxylase
4rdvB N-formimino-L-glutamate iminiohydrolase
Figure 8. Input set used in all against all comparison example.

4.2 Position of amidohydrolases and PHP superfamily in structure space

The set of structures from Figure 8 was submitted to all against all comparison, yielding the results shown
in Figure 6. You can use this live example to reproduce the figures. The dendrogram (Figure 9) and
correspondence analysis plot (Figure 10) agree quite well in grouping the most strongly similar structures.
However, branching order nearer the root becomes more or less arbitrary. For example, adenosine
deaminase and AMP deaminase (Group A) are far apart in the dendrogram but adjacent in the
correspondence analysis plot. Although members of the PHP superfamily occasionally appear in the
structural neighbour lists of amidohydrolases, the two superfamilies appear as structurally distinct in our
analyses. In the correspondence analysis plot, the first eigenvector (horizontal) separates PHP domain

9
proteins from amidohydrolases. PHP domains have a 7-stranded beta barrel, while amidohydrolases have
an 8-stranded beta barrel. The second eigenvector (vertical) separates amidohydrolases with the catalytic
and small domain from those with only the catalytic domain. The outgroup is near the origin, indicating
that it has no special affinity towards any of the other groups.

Figure 9: Dendrogram representation of protein structure space. Node labels were added manually.

10
Figure 10. Correspondence analysis plot. Groups were manually defined and correspond to those in the
dendrogram (Figure 9).

4.3 Overlapping sequence motifs of amidohydrolases and PHP superfamily

Both PHP domains and amidohydrolases bind metal ions in their active site. Figure 11 shows a stacked
sequence logo comparison of two amidohydrolases and two PHP domains. The representatives were
chosen from the extremes of the PHP cloud and amidohydrolase cloud in the correspondence analysis plot.
The idea is to bring out invariant aminodhydrolase features (conserved in both amidohydrolases) and
invariant PHP domain features (conserved in both PHP domains). We see that although some residues are
commonly conserved in both superfamilies, each has unique features missing from the other. It is
interesting that these distinct sequence motifs coincide with the structural distinction of PHP domains from
amidohydrolases (Figure 10).

Figure 11. Sequence logos for two amidohydrolases (top) and two PHP proteins (bottom).

4.4 Looking into PHP superfamily

The dendrogram (Figure 9) shows a deep divide within the PHP superfamily between phosphatases on the
one hand and phosphoesterases on the other. In the correspondence analysis plot, they are separated by
the first eigenvector (Figure 10). The structural similarity between the two subclasses dips almost as low as
the outlier. Without the unifying sequence motif, we would even question whether they are evolutionarily
related. Inspection of 3D superimpositions and multiple structural alignments revealed that while one
subsclass has a parallel alpha/beta barrel like amidohydrolases, one of the beta strands has reversed
direction in the other subclass (Figure 12). Because Dali reports only sequential alignments, this explains
why the alignment score is lower between the subclasses.

11
Figure 12. Structural conservation of the eight PHP domains in our data set. Blue regions are aligned in all
eight PHP domain structures, the green strand of the barrel has reversed direction in some.

4.5 Diversity of molecular functions in amidohydrolase superfamily

The structural dendrogram is labelled with the proteins’ functional descriptions. We can observe that, in
general, structural groupings coincide with functional categories. This observation fits well with the idea of
evolutionary continuity of structure and function. As a corollary, functions in incongruent positions in the
dendrogram should alert to possible misclassification. For example, deacetylase 2ogjA is incorrectly
annotated as dihydroorotase in PDB. Though corrected to deacetylase for this protein in Uniprot, the
incorrect annotation has spread to sequence neighbours which remain annotated as dihydroorotases in
Uniprot (as you will see in a SANS search). Our input set contains a number of uncharacterized proteins.
Functionally characterized neighbours in structure space can direct the formulation of testable hypotheses
of their molecular function, at least regarding the class of enzyme function if not precise substrate
specificity. For example, ochratoxinase and molinate hydrolase are recently evolved new enzymes.

4.6 Conclusion
The Dali server is a useful aid in structural classification. It generates overviews of selected portions of
structure space and sequence space with just a few mouse clicks. Similar narratives as above can be
developed for any structurally characterized superfamily.

5 Downloads
The DaliLite software is available for academic use from
http://ekhidna.biocenter.helsinki.fi/dali/downloads/download.html.

12
Appendix A: Sample PDB entry
HEADER PANCREATIC HORMONE 16-JAN-81 1PPT
TITLE X-RAY ANALYSIS (1.4-ANGSTROMS RESOLUTION) OF AVIAN PANCREATIC
TITLE 2 POLYPEPTIDE. SMALL GLOBULAR PROTEIN HORMONE
COMPND MOL_ID: 1;
COMPND 2 MOLECULE: AVIAN PANCREATIC POLYPEPTIDE;
COMPND 3 CHAIN: A;
COMPND 4 ENGINEERED: YES
SOURCE MOL_ID: 1;
SOURCE 2 ORGANISM_SCIENTIFIC: MELEAGRIS GALLOPAVO;
SOURCE 3 ORGANISM_COMMON: TURKEY;
SOURCE 4 ORGANISM_TAXID: 9103
AUTHOR T.L.BLUNDELL,J.E.PITTS,I.J.TICKLE,S.P.WOOD
JRNL AUTH T.L.BLUNDELL,J.E.PITTS,I.J.TICKLE,S.P.WOOD,C.W.WU
JRNL TITL X-RAY ANALYSIS (1. 4-A RESOLUTION) OF AVIAN PANCREATIC
JRNL TITL 2 POLYPEPTIDE: SMALL GLOBULAR PROTEIN HORMONE.
JRNL REF PROC.NATL.ACAD.SCI.USA V. 78 4175 1981
JRNL REFN ISSN 0027-8424
JRNL PMID 16593056
JRNL DOI 10.1073/PNAS.78.7.4175
ATOM 1 N GLY A 1 2.296 -9.636 18.253 1.00 0.00 N
ATOM 2 CA GLY A 1 1.470 -9.017 17.255 1.00 0.00 C
ATOM 3 C GLY A 1 0.448 -9.983 16.703 1.00 0.00 C
ATOM 4 O GLY A 1 0.208 -11.066 17.345 1.00 0.00 O
ATOM 5 N PRO A 2 -0.170 -9.672 15.624 1.00 0.00 N
ATOM 6 CA PRO A 2 -1.135 -10.606 14.958 1.00 0.00 C
ATOM 7 C PRO A 2 -0.376 -11.824 14.490 1.00 0.00 C
ATOM 8 O PRO A 2 0.776 -11.860 14.075 1.00 0.00 O
ATOM 9 CB PRO A 2 -1.717 -9.829 13.776 1.00 0.00 C
ATOM 10 CG PRO A 2 -0.817 -8.685 13.546 1.00 0.00 C
ATOM 11 CD PRO A 2 0.108 -8.454 14.780 1.00 0.00 C
ATOM 12 N SER A 3 -1.184 -12.918 14.566 1.00 0.00 N
ATOM 13 CA SER A 3 -0.626 -14.187 14.053 1.00 0.00 C
ATOM 14 C SER A 3 -0.642 -14.190 12.493 1.00 0.00 C
ATOM 15 O SER A 3 -1.149 -13.332 11.830 1.00 0.00 O
ATOM 16 CB SER A 3 -1.360 -15.359 14.573 1.00 0.00 C
ATOM 17 OG SER A 3 -2.655 -15.234 14.212 1.00 0.00 O
ATOM 18 N GLN A 4 0.243 -14.995 11.964 1.00 0.00 N
ATOM 19 CA GLN A 4 0.489 -14.940 10.481 1.00 0.00 C
ATOM 20 C GLN A 4 -0.766 -15.384 9.734 1.00 0.00 C
ATOM 21 O GLN A 4 -1.330 -16.452 10.019 1.00 0.00 O
ATOM 22 CB GLN A 4 1.639 -15.895 10.114 1.00 0.00 C
ATOM 23 CG GLN A 4 2.182 -15.697 8.704 1.00 0.00 C
ATOM 24 CD GLN A 4 3.315 -16.670 8.366 1.00 0.00 C
ATOM 25 OE1 GLN A 4 3.718 -16.761 7.207 1.00 0.00 O
ATOM 26 NE2 GLN A 4 3.864 -17.403 9.317 1.00 0.00 N
ATOM 27 N PRO A 5 -1.196 -14.647 8.750 1.00 0.00 N
ATOM 28 CA PRO A 5 -2.414 -14.970 8.087 1.00 0.00 C
ATOM 29 C PRO A 5 -2.264 -16.297 7.258 1.00 0.00 C
ATOM 30 O PRO A 5 -1.184 -16.595 6.819 1.00 0.00 O
ATOM 31 CB PRO A 5 -2.798 -13.854 7.153 1.00 0.00 C
ATOM 32 CG PRO A 5 -1.809 -12.748 7.438 1.00 0.00 C
ATOM 33 CD PRO A 5 -0.768 -13.190 8.408 1.00 0.00 C
ATOM 34 N THR A 6 -3.381 -16.917 7.174 1.00 0.00 N
ATOM 35 CA THR A 6 -3.548 -18.158 6.308 1.00 0.00 C
ATOM 36 C THR A 6 -3.745 -17.747 4.861 1.00 0.00 C
ATOM 37 O THR A 6 -4.693 -17.045 4.518 1.00 0.00 O
ATOM 38 CB THR A 6 -4.752 -18.911 6.884 1.00 0.00 C
ATOM 39 OG1 THR A 6 -4.040 -19.502 8.074 1.00 0.00 O
ATOM 40 CG2 THR A 6 -4.799 -20.260 6.058 1.00 0.00 C
ATOM 41 N TYR A 7 -2.893 -18.207 3.953 1.00 0.00 N
ATOM 42 CA TYR A 7 -3.065 -18.017 2.495 1.00 0.00 C
ATOM 43 C TYR A 7 -4.327 -18.738 2.010 1.00 0.00 C
ATOM 44 O TYR A 7 -4.536 -19.927 2.291 1.00 0.00 O
ATOM 45 CB TYR A 7 -1.828 -18.587 1.791 1.00 0.00 C
ATOM 46 CG TYR A 7 -1.913 -18.407 0.265 1.00 0.00 C
ATOM 47 CD1 TYR A 7 -2.029 -17.122 -0.283 1.00 0.00 C
ATOM 48 CD2 TYR A 7 -1.884 -19.519 -0.588 1.00 0.00 C
ATOM 49 CE1 TYR A 7 -2.090 -16.948 -1.671 1.00 0.00 C
ATOM 50 CE2 TYR A 7 -1.943 -19.344 -1.978 1.00 0.00 C

13
ATOM 51 CZ TYR A 7 -2.039 -18.057 -2.521 1.00 0.00 C
ATOM 52 OH TYR A 7 -2.067 -17.876 -3.868 1.00 0.00 O
ATOM 53 N PRO A 8 -5.261 -18.068 1.439 1.00 0.00 N
ATOM 54 CA PRO A 8 -6.566 -18.626 0.996 1.00 0.00 C
ATOM 55 C PRO A 8 -6.492 -19.530 -0.193 1.00 0.00 C
ATOM 56 O PRO A 8 -7.584 -20.240 -0.510 1.00 0.00 O
ATOM 57 CB PRO A 8 -7.488 -17.397 0.798 1.00 0.00 C
ATOM 58 CG PRO A 8 -6.583 -16.313 0.428 1.00 0.00 C
ATOM 59 CD PRO A 8 -5.230 -16.608 1.173 1.00 0.00 C
ATOM 60 N GLY A 9 -5.375 -19.740 -0.857 1.00 0.00 N
ATOM 61 CA GLY A 9 -5.423 -20.730 -1.983 1.00 0.00 C
ATOM 62 C GLY A 9 -5.256 -19.861 -3.214 1.00 0.00 C
ATOM 63 O GLY A 9 -5.790 -18.783 -3.338 1.00 0.00 O
ATOM 64 N ASP A 10 -4.626 -20.463 -4.248 1.00 0.00 N
ATOM 65 CA ASP A 10 -4.521 -19.865 -5.611 1.00 0.00 C
ATOM 66 C ASP A 10 -5.891 -19.644 -6.232 1.00 0.00 C
ATOM 67 O ASP A 10 -6.079 -18.696 -7.006 1.00 0.00 O
ATOM 68 CB ASP A 10 -3.697 -20.772 -6.523 1.00 0.00 C
ATOM 69 CG ASP A 10 -2.225 -20.895 -6.117 1.00 0.00 C
ATOM 70 OD1 ASP A 10 -1.521 -21.886 -6.544 1.00 0.00 O
ATOM 71 OD2 ASP A 10 -1.682 -20.014 -5.347 1.00 0.00 O
ATOM 72 N ASP A 11 -6.810 -20.490 -5.917 1.00 0.00 N
ATOM 73 CA ASP A 11 -8.106 -20.411 -6.537 1.00 0.00 C
ATOM 74 C ASP A 11 -9.141 -19.681 -5.693 1.00 0.00 C
ATOM 75 O ASP A 11 -10.273 -19.451 -6.151 1.00 0.00 O
ATOM 76 CB ASP A 11 -8.681 -21.809 -6.852 1.00 0.00 C
ATOM 77 CG ASP A 11 -7.791 -22.570 -7.829 1.00 0.00 C
ATOM 78 OD1 ASP A 11 -7.396 -21.995 -8.913 1.00 0.00 O
ATOM 79 OD2 ASP A 11 -7.431 -23.778 -7.563 1.00 0.00 O
ATOM 80 N ALA A 12 -8.612 -18.887 -4.775 1.00 0.00 N
ATOM 81 CA ALA A 12 -9.622 -18.132 -4.017 1.00 0.00 C
ATOM 82 C ALA A 12 -10.101 -16.933 -4.820 1.00 0.00 C
ATOM 83 O ALA A 12 -9.482 -16.473 -5.779 1.00 0.00 O
ATOM 84 CB ALA A 12 -8.829 -17.575 -2.793 1.00 0.00 C
ATOM 85 N PRO A 13 -11.366 -16.547 -4.687 1.00 0.00 N
ATOM 86 CA PRO A 13 -11.981 -15.406 -5.466 1.00 0.00 C
ATOM 87 C PRO A 13 -11.187 -14.121 -5.215 1.00 0.00 C
ATOM 88 O PRO A 13 -10.522 -13.958 -4.032 1.00 0.00 O
ATOM 89 CB PRO A 13 -13.424 -15.243 -4.908 1.00 0.00 C
ATOM 90 CG PRO A 13 -13.500 -16.009 -3.659 1.00 0.00 C
ATOM 91 CD PRO A 13 -12.236 -16.882 -3.480 1.00 0.00 C
ATOM 92 N VAL A 14 -11.180 -13.190 -6.126 1.00 0.00 N
ATOM 93 CA VAL A 14 -10.341 -11.949 -5.914 1.00 0.00 C
ATOM 94 C VAL A 14 -10.673 -11.235 -4.610 1.00 0.00 C
ATOM 95 O VAL A 14 -9.729 -10.720 -3.902 1.00 0.00 O
ATOM 96 CB VAL A 14 -10.477 -11.110 -7.162 1.00 0.00 C
ATOM 97 CG1 VAL A 14 -9.809 -9.750 -7.062 1.00 0.00 C
ATOM 98 CG2 VAL A 14 -10.013 -11.873 -8.431 1.00 0.00 C
ATOM 99 N GLU A 15 -11.842 -11.320 -4.113 1.00 0.00 N
ATOM 100 CA GLU A 15 -12.142 -10.553 -2.855 1.00 0.00 C
ATOM 101 C GLU A 15 -11.451 -11.174 -1.704 1.00 0.00 C
ATOM 102 O GLU A 15 -11.170 -10.380 -0.760 1.00 0.00 O
ATOM 103 CB GLU A 15 -13.711 -10.553 -2.589 1.00 0.00 C
ATOM 104 CG GLU A 15 -14.210 -11.854 -1.955 1.00 0.00 C
ATOM 105 CD GLU A 15 -15.561 -11.698 -1.254 1.00 0.00 C
ATOM 106 OE1 GLU A 15 -16.152 -12.731 -0.757 1.00 0.00 O
ATOM 107 OE2 GLU A 15 -16.108 -10.534 -1.161 1.00 0.00 O
ATOM 108 N ASP A 16 -11.319 -12.496 -1.702 1.00 0.00 N
ATOM 109 CA ASP A 16 -10.488 -13.190 -0.722 1.00 0.00 C
ATOM 110 C ASP A 16 -9.000 -12.982 -0.958 1.00 0.00 C
ATOM 111 O ASP A 16 -8.238 -12.840 0.013 1.00 0.00 O
ATOM 112 CB ASP A 16 -10.706 -14.670 -0.638 1.00 0.00 C
ATOM 113 CG ASP A 16 -12.106 -15.022 -0.152 1.00 0.00 C
ATOM 114 OD1 ASP A 16 -12.571 -16.208 -0.343 1.00 0.00 O
ATOM 115 OD2 ASP A 16 -12.821 -14.131 0.443 1.00 0.00 O
ATOM 116 N LEU A 17 -8.476 -12.788 -2.105 1.00 0.00 N
ATOM 117 CA LEU A 17 -7.028 -12.438 -2.126 1.00 0.00 C
ATOM 118 C LEU A 17 -6.810 -10.983 -1.717 1.00 0.00 C
ATOM 119 O LEU A 17 -5.812 -10.718 -1.159 1.00 0.00 O
ATOM 120 CB LEU A 17 -6.647 -12.470 -3.630 1.00 0.00 C
ATOM 121 CG LEU A 17 -6.525 -13.931 -4.064 1.00 0.00 C

14
ATOM 122 CD1 LEU A 17 -6.250 -13.916 -5.582 1.00 0.00 C
ATOM 123 CD2 LEU A 17 -5.372 -14.656 -3.263 1.00 0.00 C
ATOM 124 N ILE A 18 -7.786 -10.075 -1.877 1.00 0.00 N
ATOM 125 CA ILE A 18 -7.676 -8.682 -1.354 1.00 0.00 C
ATOM 126 C ILE A 18 -7.789 -8.754 0.184 1.00 0.00 C
ATOM 127 O ILE A 18 -6.937 -8.096 0.818 1.00 0.00 O
ATOM 128 CB ILE A 18 -8.822 -7.866 -1.960 1.00 0.00 C
ATOM 129 CG1 ILE A 18 -8.514 -7.564 -3.452 1.00 0.00 C
ATOM 130 CG2 ILE A 18 -8.973 -6.595 -1.116 1.00 0.00 C
ATOM 131 CD1 ILE A 18 -9.640 -6.945 -4.117 1.00 0.00 C
ATOM 132 N ARG A 19 -8.698 -9.540 0.771 1.00 0.00 N
ATOM 133 CA ARG A 19 -8.636 -9.644 2.284 1.00 0.00 C
ATOM 134 C ARG A 19 -7.271 -10.180 2.719 1.00 0.00 C
ATOM 135 O ARG A 19 -6.742 -9.797 3.773 1.00 0.00 O
ATOM 136 CB ARG A 19 -9.736 -10.584 2.604 1.00 0.00 C
ATOM 137 CG ARG A 19 -11.010 -9.860 3.115 1.00 0.00 C
ATOM 138 CD ARG A 19 -12.209 -10.742 3.604 1.00 0.00 C
ATOM 139 NE ARG A 19 -13.244 -10.589 2.620 1.00 0.00 N
ATOM 140 CZ ARG A 19 -14.562 -10.256 2.523 1.00 0.00 C
ATOM 141 NH1 ARG A 19 -15.649 -9.931 3.450 1.00 0.00 N
ATOM 142 NH2 ARG A 19 -14.877 -10.167 1.331 1.00 0.00 N
ATOM 143 N PHE A 20 -6.704 -11.222 2.134 1.00 0.00 N
ATOM 144 CA PHE A 20 -5.441 -11.780 2.548 1.00 0.00 C
ATOM 145 C PHE A 20 -4.374 -10.694 2.394 1.00 0.00 C
ATOM 146 O PHE A 20 -3.489 -10.541 3.246 1.00 0.00 O
ATOM 147 CB PHE A 20 -5.101 -12.968 1.646 1.00 0.00 C
ATOM 148 CG PHE A 20 -3.710 -13.511 1.946 1.00 0.00 C
ATOM 149 CD1 PHE A 20 -2.666 -13.334 1.030 1.00 0.00 C
ATOM 150 CD2 PHE A 20 -3.489 -14.185 3.148 1.00 0.00 C
ATOM 151 CE1 PHE A 20 -1.392 -13.833 1.327 1.00 0.00 C
ATOM 152 CE2 PHE A 20 -2.218 -14.680 3.443 1.00 0.00 C
ATOM 153 CZ PHE A 20 -1.168 -14.502 2.535 1.00 0.00 C
ATOM 154 N TYR A 21 -4.347 -9.918 1.280 1.00 0.00 N
ATOM 155 CA TYR A 21 -3.402 -8.885 1.007 1.00 0.00 C
ATOM 156 C TYR A 21 -3.411 -7.902 2.181 1.00 0.00 C
ATOM 157 O TYR A 21 -2.361 -7.533 2.718 1.00 0.00 O
ATOM 158 CB TYR A 21 -3.821 -8.142 -0.271 1.00 0.00 C
ATOM 159 CG TYR A 21 -3.056 -6.870 -0.555 1.00 0.00 C
ATOM 160 CD1 TYR A 21 -1.730 -6.925 -0.993 1.00 0.00 C
ATOM 161 CD2 TYR A 21 -3.708 -5.654 -0.376 1.00 0.00 C
ATOM 162 CE1 TYR A 21 -1.042 -5.733 -1.246 1.00 0.00 C
ATOM 163 CE2 TYR A 21 -3.022 -4.467 -0.625 1.00 0.00 C
ATOM 164 CZ TYR A 21 -1.692 -4.505 -1.058 1.00 0.00 C
ATOM 165 OH TYR A 21 -1.035 -3.335 -1.287 1.00 0.00 O
ATOM 166 N ASP A 22 -4.642 -7.469 2.585 1.00 0.00 N
ATOM 167 CA ASP A 22 -4.728 -6.473 3.688 1.00 0.00 C
ATOM 168 C ASP A 22 -4.207 -7.089 4.978 1.00 0.00 C
ATOM 169 O ASP A 22 -3.568 -6.404 5.780 1.00 0.00 O
ATOM 170 CB ASP A 22 -6.194 -6.063 3.854 1.00 0.00 C
ATOM 171 CG ASP A 22 -6.707 -5.158 2.746 1.00 0.00 C
ATOM 172 OD1 ASP A 22 -7.967 -5.120 2.494 1.00 0.00 O
ATOM 173 OD2 ASP A 22 -5.890 -4.431 2.071 1.00 0.00 O
ATOM 174 N ASN A 23 -4.505 -8.385 5.271 1.00 0.00 N
ATOM 175 CA ASN A 23 -4.038 -8.947 6.542 1.00 0.00 C
ATOM 176 C ASN A 23 -2.506 -9.142 6.402 1.00 0.00 C
ATOM 177 O ASN A 23 -1.837 -8.942 7.451 1.00 0.00 O
ATOM 178 CB ASN A 23 -4.716 -10.254 6.855 1.00 0.00 C
ATOM 179 CG ASN A 23 -6.186 -10.108 7.257 1.00 0.00 C
ATOM 180 OD1 ASN A 23 -6.841 -11.277 7.242 1.00 0.00 O
ATOM 181 ND2 ASN A 23 -6.691 -9.038 7.581 1.00 0.00 N
ATOM 182 N LEU A 24 -1.987 -9.505 5.267 1.00 0.00 N
ATOM 183 CA LEU A 24 -0.451 -9.679 5.213 1.00 0.00 C
ATOM 184 C LEU A 24 0.173 -8.275 5.430 1.00 0.00 C
ATOM 185 O LEU A 24 1.220 -8.264 5.981 1.00 0.00 O
ATOM 186 CB LEU A 24 -0.202 -10.148 3.805 1.00 0.00 C
ATOM 187 CG LEU A 24 1.277 -10.481 3.548 1.00 0.00 C
ATOM 188 CD1 LEU A 24 1.764 -11.470 4.539 1.00 0.00 C
ATOM 189 CD2 LEU A 24 1.499 -11.012 2.076 1.00 0.00 C
ATOM 190 N GLN A 25 -0.410 -7.204 4.911 1.00 0.00 N
ATOM 191 CA GLN A 25 0.085 -5.877 5.096 1.00 0.00 C
ATOM 192 C GLN A 25 0.191 -5.541 6.582 1.00 0.00 C

15
ATOM 193 O GLN A 25 1.265 -5.150 7.065 1.00 0.00 O
ATOM 194 CB GLN A 25 -0.806 -4.832 4.429 1.00 0.00 C
ATOM 195 CG GLN A 25 -0.281 -3.402 4.489 1.00 0.00 C
ATOM 196 CD GLN A 25 -0.921 -2.504 3.422 1.00 0.00 C
ATOM 197 OE1 GLN A 25 -0.397 -1.428 3.134 1.00 0.00 O
ATOM 198 NE2 GLN A 25 -2.028 -2.888 2.811 1.00 0.00 N
ATOM 199 N GLN A 26 -0.856 -5.749 7.310 1.00 0.00 N
ATOM 200 CA GLN A 26 -0.856 -5.563 8.757 1.00 0.00 C
ATOM 201 C GLN A 26 0.308 -6.289 9.398 1.00 0.00 C
ATOM 202 O GLN A 26 1.009 -5.921 10.266 1.00 0.00 O
ATOM 203 CB GLN A 26 -2.266 -5.866 9.301 1.00 0.00 C
ATOM 204 CG GLN A 26 -2.357 -5.747 10.824 1.00 0.00 C
ATOM 205 CD GLN A 26 -2.333 -4.297 11.313 1.00 0.00 C
ATOM 206 OE1 GLN A 26 -2.414 -4.053 12.516 1.00 0.00 O
ATOM 207 NE2 GLN A 26 -2.225 -3.309 10.444 1.00 0.00 N
ATOM 208 N TYR A 27 0.366 -7.626 9.157 1.00 0.00 N
ATOM 209 CA TYR A 27 1.356 -8.520 9.698 1.00 0.00 C
ATOM 210 C TYR A 27 2.759 -8.036 9.359 1.00 0.00 C
ATOM 211 O TYR A 27 3.622 -7.942 10.245 1.00 0.00 O
ATOM 212 CB TYR A 27 1.111 -9.949 9.116 1.00 0.00 C
ATOM 213 CG TYR A 27 2.122 -10.986 9.594 1.00 0.00 C
ATOM 214 CD1 TYR A 27 3.124 -11.439 8.729 1.00 0.00 C
ATOM 215 CD2 TYR A 27 2.044 -11.485 10.899 1.00 0.00 C
ATOM 216 CE1 TYR A 27 4.061 -12.377 9.173 1.00 0.00 C
ATOM 217 CE2 TYR A 27 2.984 -12.421 11.346 1.00 0.00 C
ATOM 218 CZ TYR A 27 3.998 -12.862 10.486 1.00 0.00 C
ATOM 219 OH TYR A 27 4.932 -13.751 10.922 1.00 0.00 O
ATOM 220 N LEU A 28 3.051 -7.735 8.171 1.00 0.00 N
ATOM 221 CA LEU A 28 4.342 -7.279 7.784 1.00 0.00 C
ATOM 222 C LEU A 28 4.675 -5.907 8.478 1.00 0.00 C
ATOM 223 O LEU A 28 5.887 -5.811 8.927 1.00 0.00 O
ATOM 224 CB LEU A 28 4.614 -7.138 6.312 1.00 0.00 C
ATOM 225 CG LEU A 28 4.490 -8.512 5.562 1.00 0.00 C
ATOM 226 CD1 LEU A 28 4.429 -8.253 3.969 1.00 0.00 C
ATOM 227 CD2 LEU A 28 5.761 -9.362 5.838 1.00 0.00 C
ATOM 228 N ASN A 29 3.737 -4.993 8.530 1.00 0.00 N
ATOM 229 CA ASN A 29 4.064 -3.735 9.328 1.00 0.00 C
ATOM 230 C ASN A 29 4.502 -4.080 10.751 1.00 0.00 C
ATOM 231 O ASN A 29 5.252 -3.321 11.381 1.00 0.00 O
ATOM 232 CB ASN A 29 2.748 -2.894 9.387 1.00 0.00 C
ATOM 233 CG ASN A 29 2.581 -2.132 8.083 1.00 0.00 C
ATOM 234 OD1 ASN A 29 1.565 -1.465 7.896 1.00 0.00 O
ATOM 235 ND2 ASN A 29 3.539 -2.200 7.175 1.00 0.00 N
ATOM 236 N VAL A 30 3.812 -5.044 11.456 1.00 0.00 N
ATOM 237 CA VAL A 30 4.069 -5.352 12.824 1.00 0.00 C
ATOM 238 C VAL A 30 5.379 -6.029 12.954 1.00 0.00 C
ATOM 239 O VAL A 30 6.270 -5.680 13.834 1.00 0.00 O
ATOM 240 CB VAL A 30 2.961 -6.250 13.460 1.00 0.00 C
ATOM 241 CG1 VAL A 30 3.526 -6.593 14.954 1.00 0.00 C
ATOM 242 CG2 VAL A 30 1.683 -5.435 13.674 1.00 0.00 C
ATOM 243 N VAL A 31 5.698 -6.965 12.038 1.00 0.00 N
ATOM 244 CA VAL A 31 7.001 -7.699 12.162 1.00 0.00 C
ATOM 245 C VAL A 31 8.157 -6.782 11.957 1.00 0.00 C
ATOM 246 O VAL A 31 9.302 -7.004 12.397 1.00 0.00 O
ATOM 247 CB VAL A 31 6.845 -8.810 10.979 1.00 0.00 C
ATOM 248 CG1 VAL A 31 8.217 -9.440 10.757 1.00 0.00 C
ATOM 249 CG2 VAL A 31 5.842 -9.913 11.342 1.00 0.00 C
ATOM 250 N THR A 32 8.037 -5.850 11.015 1.00 0.00 N
ATOM 251 CA THR A 32 9.068 -4.861 10.736 1.00 0.00 C
ATOM 252 C THR A 32 8.975 -3.622 11.693 1.00 0.00 C
ATOM 253 O THR A 32 9.882 -2.752 11.522 1.00 0.00 O
ATOM 254 CB THR A 32 8.833 -4.336 9.342 1.00 0.00 C
ATOM 255 OG1 THR A 32 7.762 -3.572 9.058 1.00 0.00 O
ATOM 256 CG2 THR A 32 9.614 -4.848 8.439 1.00 0.00 C
ATOM 257 N ARG A 33 8.154 -3.603 12.666 1.00 0.00 N
ATOM 258 CA ARG A 33 7.995 -2.499 13.658 1.00 0.00 C
ATOM 259 C ARG A 33 7.787 -1.166 12.943 1.00 0.00 C
ATOM 260 O ARG A 33 8.043 -0.093 13.507 1.00 0.00 O
ATOM 261 CB ARG A 33 9.235 -2.379 14.561 1.00 0.00 C
ATOM 262 CG ARG A 33 9.711 -3.734 15.065 1.00 0.00 C
ATOM 263 CD ARG A 33 10.875 -3.613 16.039 1.00 0.00 C

16
ATOM 264 NE ARG A 33 10.549 -2.811 17.223 1.00 0.00 N
ATOM 265 CZ ARG A 33 9.938 -3.306 18.303 1.00 0.00 C
ATOM 266 NH1 ARG A 33 9.575 -4.596 18.352 1.00 0.00 N
ATOM 267 NH2 ARG A 33 9.649 -2.584 19.395 1.00 0.00 N
ATOM 268 N HIS A 34 7.031 -1.228 11.897 1.00 0.00 N
ATOM 269 CA HIS A 34 6.779 0.039 11.099 1.00 0.00 C
ATOM 270 C HIS A 34 5.289 0.163 10.798 1.00 0.00 C
ATOM 271 O HIS A 34 4.835 -0.137 9.689 1.00 0.00 O
ATOM 272 CB HIS A 34 7.587 -0.011 9.878 1.00 0.00 C
ATOM 273 CG HIS A 34 7.608 1.293 9.098 1.00 0.00 C
ATOM 274 ND1 HIS A 34 6.953 1.430 7.879 1.00 0.00 N
ATOM 275 CD2 HIS A 34 8.195 2.486 9.363 1.00 0.00 C
ATOM 276 CE1 HIS A 34 7.144 2.668 7.454 1.00 0.00 C
ATOM 277 NE2 HIS A 34 7.884 3.310 8.330 1.00 0.00 N
ATOM 278 N ARG A 35 4.524 0.544 11.879 1.00 0.00 N
ATOM 279 CA ARG A 35 3.108 0.616 11.852 1.00 0.00 C
ATOM 280 C ARG A 35 2.637 1.882 11.134 1.00 0.00 C
ATOM 281 O ARG A 35 1.446 2.229 11.173 1.00 0.00 O
ATOM 282 CB ARG A 35 2.550 0.636 13.314 1.00 0.00 C
ATOM 283 CG ARG A 35 2.848 -0.712 13.994 1.00 0.00 C
ATOM 284 CD ARG A 35 2.475 -0.788 15.476 1.00 0.00 C
ATOM 285 NE ARG A 35 3.312 -1.745 16.223 1.00 0.00 N
ATOM 286 CZ ARG A 35 2.837 -2.773 16.945 1.00 0.00 C
ATOM 287 NH1 ARG A 35 1.521 -3.007 17.037 1.00 0.00 N
ATOM 288 NH2 ARG A 35 3.612 -3.634 17.621 1.00 0.00 N
ATOM 289 N TYR A 36 3.365 2.609 10.443 1.00 0.00 N
ATOM 290 CA TYR A 36 2.765 3.446 9.296 1.00 0.00 C
ATOM 291 C TYR A 36 2.332 2.479 8.197 1.00 0.00 C
ATOM 292 O TYR A 36 3.166 1.720 7.671 1.00 0.00 O
ATOM 293 CB TYR A 36 4.021 4.330 8.787 1.00 0.00 C
ATOM 294 CG TYR A 36 4.734 4.795 10.058 1.00 0.00 C
ATOM 295 CD1 TYR A 36 5.675 3.963 10.681 1.00 0.00 C
ATOM 296 CD2 TYR A 36 4.424 6.040 10.616 1.00 0.00 C
ATOM 297 CE1 TYR A 36 6.332 4.393 11.840 1.00 0.00 C
ATOM 298 CE2 TYR A 36 5.083 6.471 11.773 1.00 0.00 C
ATOM 299 CZ TYR A 36 6.043 5.652 12.379 1.00 0.00 C
ATOM 300 OH TYR A 36 6.704 6.088 13.485 1.00 0.00 O
ATOM 301 OXT TYR A 36 1.276 2.139 7.885 1.00 0.00 O
HETATM 303 ZN ZN A 37 1.119 -11.175 19.270 1.00 0.00 ZN

17
Appendix B: Input data for plot in Figure 10
PDBID EV1 A B C D E F G H Outlier

1a4mA 0.04 -0.02

2a3lA 0.03 0.00

1j5sA -0.06 0.21

2qpxA -0.03 0.15

3iacA -0.07 0.21

1bf6A -0.02 0.08

2ob3A -0.01 0.09

2vc5A -0.02 0.09

2y1hB -0.01 0.06

3gg7A -0.03 0.05

3k2gB -0.02 0.09

1itqA -0.04 0.13

2dvtA -0.07 0.22

2ffiA -0.04 0.12

2gwgA -0.05 0.21

3cjpA -0.04 0.12

3irsA -0.03 0.14

4dlfA -0.05 0.14

4dziC -0.05 0.21

4hk5D -0.08 0.22

4mupB -0.05 0.11

4ofcA -0.07 0.23

4qrnA -0.07 0.25

1a5kC 0.09 0.01

1gkpA 0.08 0.00

1onxA 0.09 -0.04

1yrrB 0.08 -0.05

2vunA 0.09 -0.05

3e74A 0.10 -0.02

3giqA 0.10 -0.02

3griA 0.09 -0.01

3nqbA 0.09 -0.06

3pnuA 0.01 0.08

4b3zD 0.08 0.02

1j6pA 0.14 -0.13

18
1k6wA 0.10 -0.09

2imrA 0.07 -0.07

2ogjA 0.10 -0.04

2oofA 0.11 -0.11

2pajA 0.13 -0.11

2uz9A 0.14 -0.13

3icjA 0.11 -0.08

3ls9A 0.13 -0.12

3mkvA 0.12 -0.10

3mtwA 0.12 -0.10

3ooqA 0.13 -0.09

4c5yA 0.12 -0.10

4cqbA 0.11 -0.09

4rdvB 0.14 -0.13

4ub9A 0.11 -0.08

4v1xE 0.14 -0.12

1m65A -0.18 -0.09

1v77A -0.04 -0.06

3au2A -0.20 -0.07

3dcpA -0.15 -0.08

3qy6A -0.08 0.02

2anuA -0.35 -0.23

2yb1A -0.38 -0.26

3e38A -0.46 -0.32

3f2bA -0.20 -0.13

1bksA -0.06 -0.01

19