Access augmentation or optimizing

Method for clustering sequences in groups

6304868

Abstract

In order to cluster sequences to biological groups, the conventional databank search programs are iteratively called in with a view to clustering various related sequences to one determined protein sequence. The inventive method enables full automatic distribution of a high number of protein sequences in groups. The major part of such groups are segregated, so that they represent a meaningful and valid grouping of data.


Claims

What is claimed is:

1. A method of grouping sequences in a sequence database to provide a sequence cluster containing similar sequences, the method comprising:

(a) providing an inquiry sequence;

(b) determining a positive quantity of similar sequences to said inquiry sequence from the sequence database using a database search program, wherein said positive quantity of similar sequences includes all sequences in said sequence database for which the probability that similarity occurs randomly is below a predetermined threshold level;

(c) selecting one sequence having the highest probability of randomly occurred similarity from said positive quantity determined for said inquiry sequence as a subsequent search sequence;

(d) determining a subsequent positive quantity of similar sequences for said subsequent search sequence from the sequence database using the database search program as in (b);

(e) identifing new sequences in said subsequent positive quantity that are not included in said positive quantity determined for said inquiry sequence, said new sequences being combined with said positive quantity of similar sequences of said inquiry sequence;

(f) repeating steps (c) to (e) as long as new sequences are still identified that are not contained in the combined positive quantity of similar sequences of step (e) and as long as there is an intersection between sequences in the combined positive quantity and any sequences in the subsequent positive quantity; and

(g) outputting all the different sequences contained in the combined positive quantity as one cluster containing similar sequences.

2. A method of clustering all sequences in a sequence database, according to the method of claim 1, further comprising:

(h) repeating steps (a)-(g) for each sequence in the sequence database to obtain multiple clusters,

(i) removing clusters that occur more than once except for one cluster;

(j) removing clusters that are contained in other clusters,

(j) of the remaining clusters, outputting those clusters that do not overlap with other clusters as partitioning of the database, and

(k) outputting remaining overlapping clusters as groups whose clusters are linked together by overlapping.

3. The method of claim 1, wherein the threshold level of probability is between 10.sup.-20 and 10.sup.-35.

4. The method of claim 2, wherein the threshold level of probability is between 10.sup.-20 and 10.sup.-35.


Description

This invention concerns a method of grouping sequences in families.

Large quantities of protein sequence data are generated today in molecular biology. A major problem here is how to group such protein sequence data logically in biological families. Since families are not defined exactly, but instead the diversity of different gene families varies, this involves a problem in data grouping which is not at all trivial.

In the past, biological information could only be of assistance for human experts who would thoroughly research the output of database searching programs and would create a grouping according to families. This method is time-consuming, labor-intensive and not very reproducible.

Therefore, the object of this invention is to find a method with which a large set of protein sequences can be divided into groups fully automatically.

This object is achieved with the features of Patent Claim 1.

The method described here is based on the finding that rapid grouping can be achieved when traditional database searching programs are run iteratively to find a quantity of sequences related to a given protein sequence.

It is advantageous if the method described here is carried out for each sequence in the database, removing clusters that occur repeatedly except for one cluster each, removing clusters that are contained in other clusters; of the remaining quantity of clusters, outputting the clusters that do not overlap with other clusters as partitioning of the database and outputting the remaining portion of the overlapping clusters as groups whose clusters are linked together by overlapping.

In this way, database clustering is achieved, resulting in the fact that most of the clusters are disjunctive to one another, and therefore a valid and reasonable clustering of data is achieved.

This method permits a much faster and more objective analysis of new sequence data than has been possible in the past. The paired disjunctive part of the clustering no longer requires any checking from a practical standpoint, and it forms the ideal basis for automatic annotations and more extensive analyses. The residue, i.e., the remaining portion of overlapping clusters, is the portion that must be studied by human experts. This portion is extremely reduced and is also prestructured due to the overlapping clusters.

This method can be carried out in an extremely short computation time because it is no longer necessary to compare each sequence separately with every other sequence in order to cluster an entire database.

In an advantageous embodiment, the threshold value is between 10.sup.-20 and 10.sup.-35. In practice, a value of 10.sup.-30 has proven feasible. In addition to clustering protein sequences, this method is also suitable for DNA sequences, where it may be appropriate to relax the threshold value beyond 10.sup.-20.

A further refinement of the method according to this invention consists of the fact that of the positive quantity of sequences found in one iteration step, not only the sequence weighted as worst, is used for another database search, but also all sequences of this quantity serve as a search sequence in additional searches. Due to the larger number of searches to be performed, this alternative is not as fast as the method described originally for an individual search. In conjunction with clustering, however, this does not cause any time loss. However, since the sequence space around the initial sequence is searched more thoroughly, the resulting cluster has a high probability of already including all the sequences belonging to this protein family.

The "BLASTP" program is very suitable as a database searching program. This program is described in greater detail by S. F. Altschul, W. Gish, W. Miller, E. W. Myers, and D. J. Lipman, "Basic Local Alignment Search Tool," J. Mol. Biol., 215: 403-410, 1990.

As an alternative, however, the "FASTA" database search program may also be used as described, for example, in the following literature citation: W. R. Pearson and D. J. Lipman, "Improved tools for biological sequence comparison," Proc. Natl. Acad. Sci. USA, 85: 2444-2448, 1988.

In addition to the "BLASTP" and "FASTA" database search programs, any other database search program may also be used. For example, the "gapped BLAST" program (described by S. F. Altschul, T. L. Madden, A. A. Schaeffer, J. Zhang, Z. Zhang, W. Miller and D. J. Lipman, "Gapped BLAST and PSI-BLAST: a new generation of protein database search programs," Nucleic Acids Research, 25 (17): 3389-3402, 1997, is also suitable.

Using the "BLASTP" program, the PIR1 database, release 51, has been clustered by quantitative theory. This database contains 13,489 protein sequences and is described in detail by David G. George, Richard J. Dodson, John S. Garavelli, Daniel H. Haft, Lois T. Hunt, Christopher R. Marzee, Bruce C. Orcutt, Kathryn E. Sidman, Geetha Y. Srinivasarao, Lai-Su L. Yeh, Lieslie M. Arminski, Robert S. Ledley, Akira Tsugita and Winoma Barker, "The Protein Information Resource (PIR) and the PIR international protein sequence database," Nucleic Acids Research, 25 (1): 24-27, 1997. Grouping this database required approximately one day of computation time, and 91% of the database sequences were grouped into disjunctive clusters in a fully automatic procedure. The residue includes only approximately 9% of the database sequences.

The SWISS-PROT database, whose 34.sup.th release contains 59,021 sequences, is a much larger database. This database is described by Amos Bairoch and Rolf Apweiler, "The SWISS-PROT protein sequence data bank and its supplement TrEMBL," Nucleic Acids Research, 25 (1):31-36, 1997. Within approximately five days of computation time, 80% of the sequences were classified in disjunctive classes.

These examples show that extreme savings in terms of computation time are possible with the method according to this invention in comparison with traditional methods, and the cluster results are excellent.

An algorithm for the method described here is presented below:

cluster.rarw.empty quantity

search sequence.rarw.inquiry sequence

as long as (search sequence defined)

database searched with the search sequence

positive quantity.rarw.all found sequences below the threshold level with a probability value

search sequence.rarw.undefined

if (first search) then

reference quantity.rarw.positive quantity

end if

if (sequence exists in a positive quantity that is not contained in the cluster) and

(intersecting quantity between the positive quantity and reference quantity is not blank)

then

search sequence.rarw.sequence weighted as worst in the positive quantity not contained in the cluster

cluster.rarw.combined quantity of cluster and positive quantity

end if

end as long as

The sequence with which the search method begins is called the initial sequence, and the quantity of found sequences is called the cluster belonging to this initial sequence. First, a database search program such as "BLASTP" or "FASTA" is started with the initial sequence, and all sequences from the database that have a significant similarity with the initial sequence are accepted. We call this quantity of sequences a positive quantity, and we include it in the cluster as related sequences. There is a significant similarity between two sequences when the probability that this similarity occurs randomly is very low, i.e., below a given threshold. From the positive quantity thus obtained, we now use the sequence weighted as worst (i.e., the sequence having the highest probability) as a search sequence for another database search. This process is repeated as long as sequences below the threshold value are found which are not contained in the cluster and as long as there is an intersection quantity between the positive quantity of the initial sequence and the positive quantity of the instantaneous search sequence.

Then the following algorithm is carried out:

input: database

output: partitioning of the database and groups with overlapping clusters

perform the above database search method for all (sequences in the database)

cluster quantity.rarw.all clusters generated for all (identical clusters in the cluster quantity) remove identical clusters except for one representative

cluster quantity.rarw.clusters without identical clusters for all (clusters in the cluster quantity contained completely in another cluster) remove the smaller cluster of this cluster pair

cluster quantity.rarw.clusters without identical clusters and without inclusions

partitioning.rarw.all clusters in the cluster quantity that do not overlap

overlapping.rarw.all overlapping clusters in the cluster quantity are combined in groups

To obtain database clustering, the method described above is carried out for all sequences in the database, i.e., each sequence is assigned a cluster of sequences to which it is related. From this quantity of clusters, all identical clusters are removed except for one example, because they do not contain any additional information. The remaining cluster quantity is then examined for inclusions, and clusters that are contained completely in other clusters are removed until no more inclusions are present. Of this cluster quantity, the clusters that do not overlap with others can now be regarded as a logical partitioning of the database. The remaining small number of overlapping clusters are combined in groups whose clusters are linked together among one another by overlapping.

An embodiment of this invention is described in greater detail below.

We use the sequence of the human homeobox engrailed-1 protein (HME1_HUMAN) as the inquiry sequence and then search the Swissprot database (release 34) with it for related sequences. The search is performed with the BLASTP program, and we select a threshold with a probability of 10.sup.-30. The result of this search looks approximately as follows (excerpts): 

    Sequences found:
     Probability:
    SPR.vertline.Q05925.vertline.HME1_HUMAN HOMEOBOX PROTEIN ENGRAILED-1 (HU-E
     . . . 2.4e-279
    SPR.vertline.P09065.vertline.HME1_MOUSE HOMEOBOX PROTEIN ENGRAILED-1 (MO-E
     . . . 4.7e-189
    SPR.vertline.Q05916.vertline.HME1_CHICK HOMEOBOX PROTEIN ENGRAILED-1 (GG-E
     . . . 4.6e-132
    SPR.vertline.Q05917.vertline.HME2_CHICK HOMEOBOX PROTEIN ENGRAILED-2 (GG-E
     . . . 3.6e-95
    SPR.vertline.P19622.vertline.HME2_HUMAN HOMEOBOX PROTEIN ENGRAILED-2 (HU-E
     . . . 8.0e-95
    SPR.vertline.P09066.vertline.HME2_MOUSE HOMEOBOX PROTEIN ENGRAILED-2 (MO-E
     . . . 5.6e-92
    SPR.vertline.P09015.vertline.HME2_BRARE HOMEOBOX PROTEIN ENGRAILED-2 (ZF-E
     . . . 5.2e-70
    SPR.vertline.P31538.vertline.HMEB_XENLA HOMEOBOX PROTEIN ENGRAILED-1B (EN-
     . . . 1.5e-x
    SPR.vertline.P52729.vertline.HMEC_XENLA HOMEOBOX PROTEIN ENGRAILED-2A (EN-
     . . . 2.1e-66
    SPR.vertline.P52730.vertline.HMED_XENLA HOMEOBOX PROTEIN ENGRAILED-2B (EN-
     . . . 2.1e-65
    SPR.vertline.P31533.vertline.HME3_BRARE HOMEOBOX PROTEIN ENGRAILED-3 (ZF-E
     . . . 6.1e-64
    SPR.vertline.Q04896.vertline.HME1_BRARE HOMEOBOX PROTEIN ENGRAILED-1
     1.0e-61
    SPR.vertline.P09145.vertline.HMEN_DROVI SEGMENTATION POLARITY PROTEIN ENGR
     . . . 9.1e-59
    SPR.vertline.P05527.vertline.HMIN_DROME INVECTED PROTEIN.
     4.5e-57
    SPR.vertline.P27609.vertline.HMEN_BOMMO SEGMENTATION POLARITY PROTEIN ENGR
     . . . 1.5e-55
    SPR.vertline.P27610.vertline.HMIN_BOMMO INVECTED PROTEIN.
     1.1e-52
    SPR.vertline.P09532.vertline.HMEN_TRIGR HOMEOBOX PROTEIN ENGRAILED (SU-HB .
     . . 4.0e-44
    SPR.vertline.Q05640.vertline.HMEN_ARTSF HOMEOBOX PROTEIN ENGRAILED.
     1.7e-42
    SPR.vertline.P09076.vertline.HME3_APIME HOMEOBOX PROTEIN E30 (FRAGMENT).
     2.1e-41
    SPR.vertline.P09075.vertline.HME6_APIME HOMEOBOX PROTEIN E60 (FRAGMENT).
     1.0e-40
    SPR.vertline.P14150.vertline.HMEN_SCHAM HOMEOBOX PROTEIN ENGRAILED (G-EN .
     . . 2.3e-40
    SPR.vertline.P23397.vertline.HMEN_HELTR HOMEOBOX PROTEIN HT-EN (FRAGMENT).
     1.3e-38
    SPR.vertline.P31537.vertline.HMEA_XENLA HOMEOBOX PROTEIN ENGRAILED-1A (EN-
     . . . 7.9e-33
    SPR.vertline.P31535.vertline.HMEA_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE A .
     . . 1.1e-27
    SPR.vertline.P34326.vertline.HM16_CAEEL HOMEOBOX PROTEIN ENGRAILED-LIKE CE
     . . . 7.1e-27
    SPR.vertline.P31536.vertline.HMEB_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE B .
     . . 5.0e-26


On the basis of the threshold at 10.sup.-30, our cluster now contains the following sequences: 

    HME1_HUMAN,       HME1_MOUSE,       HME1_CHICK,     HME2_CHICK,
     HME2_HUMAN,
    HME2_MOUSE,       HME2_BRARE,       HMEB_XENLA,     HMEC_XENLA,
     HMED_XENLA,
    HME3_BRARE,       HME1_BRARE,       HMEN_DROVI,     HMIN_DROME,
     HMEN_BOMMO,
    HMEN_DROME,       HMIN_BOMMO,       HMEN_TRIGR,     HMEN_ARTSF,
     HME3_APIME,
    HME6_APIME,       HMEN_SCHAM,       HMEN_HELTR,     HMEA_XENLA.


The next run through the BLASTP program is then carried out with the sequence weighted as worst in this quantity, namely with the engrailed-1A homeobox protein of the horned toad (HMEA_XENLA). The result of this search looks as follows (excerpts): 

    Sequences found:
     Probability:
    SPR.vertline.P31538.vertline.HMEB_XENLA HOMEOBOX PROTEIN ENGRAILED-1B (EN-
     . . . 2.8e-36
    SPR.vertline.P31537.vertline.HMEA_XENLA HOMEOBOX PROTEIN ENGRAILED-1A (EN-
     . . . 3.2e-36
    SPR.vertline.P09015.vertline.HME2_BRARE HOMEOBOX PROTEIN ENGRAILED-2 (ZF-E
     . . . 1.1e-34
    SPR.vertline.Q05925.vertline.HME1_HUMAN HOMEOBOX PROTEIN ENGRAILED-1 (HU-E
     . . . 1.3e-33
    SPR.vertline.P09065.vertline.HME1_MOUSE HOMEOBOX PROTEIN ENGRAILED-1 (MO-E
     . . . 1.4e-33
    SPR.vertline.Q05916.vertline.HME1_CHICK HOMEOBOX PROTEIN ENGRAILED-1 (GG-E
     . . . 1.5e-33
    SPR.vertline.P52729.vertline.HMEC_XENLA HOMEOBOX PROTEIN ENGRAILED-2A (EN-
     . . . 9.9e-33
    SPR.vertline.P52730.vertline.HMED_XENLA HOMEOBOX PROTEIN ENGRAILED-2B (EN-
     . . . 5.9e-32
    SPR.vertline.Q05917.vertline.HME2_CHICK HOMEOBOX PROTEIN ENGRAILED-2 (GG-E
     . . . 1.3e-31
    SPR.vertline.P09066.vertline.HME2_MOUSE HOMEOBOX PROTEIN ENGRAILED-2 (MO-E
     . . . 1.8e-31
    SPR.vertline.P19622.vertline.HME2_HUMAN HOMEOBOX PROTEIN ENGRAILED-2 (HU-E
     . . . 2.0e-31
    SPR.vertline.Q04896.vertline.HME1_BRARE HOMEOBOX PROTEIN ENGRAILED-1.
     8.1e-31
    SPR.vertline.P31535.vertline.HMEA_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE A .
     . . 4.3e-30
    SPR.vertline.P31533.vertline.HME3_BRARE HOMEOBOX PROTEIN ENGRAILED-3 (ZF-E
     . . . 6.7e-30
    SPR.vertline.P31536.vertline.HMEB_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE B .
     . . 1.8e-28
    SPR.vertline.P09532.vertline.HMEN_TRIGR HOMEOBOX PROTEIN ENGRAILED (SU-HB-
     . . . 8.8e-28
    SPR.vertline.P31534.vertline.HMEN_LAMPL HOMEOBOX PROTEIN ENGRAILED-LIKE (E
     . . . 8.8e-28
    SPR.vertline.P09075.vertline.HME6_APIME HOMEOBOX PROTEIN E60 (FRAGMENT).
     2.1e-26
    SPR.vertline.P23397.vertline.HMEN_HELTR HOMEOBOX PROTEIN HT-EN (FRAGMENT).
     2.3e-26
    SPR.vertline.P09076.vertline.HME3_APIME HOMEOBOX PROTEIN E30 (FRAGMENT).
     3.9e-26


Let us again consider all sequences having a probability lower than 10.sup.-30 we and find that except for HMEA_MYXGL, all sequences are contained in the cluster. This sequence is now included in the cluster, and the next BLASTP search is started with it. This search yields the following result (excerpts): 

    Sequences found:
     Probability:
    SPR.vertline.P31535.vertline.HMEA_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE A .
     . . 3.8e-36
    SPR.vertline.P31534.vertline.HMEN_LAMPL HOMEOBOX PROTEIN ENGRAILED-LIKE (E
     . . . 1.5e-30
    SPR.vertline.P31538.vertline.HMEB_XENLA HOMEOBOX PROTEIN ENGRAILED-1B (EN-
     . . . 1.8e-30
    SPR.vertline.P31537.vertline.HMEA_XENLA HOMEOBOX PROTEIN ENGRAILED-1A (EN-
     . . . 3.8e-30
    SPR.vertline.P52729.vertline.HMEC_XENLA HOMEOBOX PROTEIN ENGRAILED-2A (EN-
     . . . 4.9e-29
    SPR.vertline.P09015.vertline.HME2_BRARE HOMEOBOX PROTEIN ENGRAILED-2 (ZF-E
     . . . 1.4e-28
    SPR.vertline.Q05925.vertline.HME1_HUMAN HOMEOBOX PROTEIN ENGRAILED-1 (HU-E
     . . . 1.7e-28
    SPR.vertline.P09065.vertline.HME1_MOUSE HOMEOBOX PROTEIN ENGRAILED-1 (MO-E
     . . . 1.8e-28
    SPR.vertline.P09066.vertline.HME2_MOUSE HOMEOBOX PROTEIN ENGRAILED-2 (MO-E
     . . . 3.1e-28
    SPR.vertline.P19622.vertline.HME2_HUMAN HOMEOBOX PROTEIN ENGRAILED-2 (HU-E
     . . . 3.3e-28
    SPR.vertline.Q05916.vertline.HME1_CHICK HOMEOBOX PROTEIN ENGRAILED-1 (GG-E
     . . . 4.6e-28
    SPR.vertline.P52730.vertline.HMED_XENLA HOMEOBOX PROTEIN ENGRAILED-2B (EN-
     . . . 2.1e-27
    SPR.vertline.Q05917.vertline.HME2_CHICK HOMEOBOX PROTEIN ENGRAILED-2 (GG-E
     . . . 2.2e-27
    SPR.vertline.P09075.vertline.HME6_APIME HOMEOBOX PROTEIN E60 (FRAGMENT).
     2.9e-27
    SPR.vertline.P23397.vertline.HMEN_HELTR HOMEOBOX PROTEIN HT-EN (FRAGMENT).
     4.4e-27
    SPR.vertline.Q04896.vertline.HME1_BRARE HOMEOBOX PROTEIN ENGRAILED-1.
     4.9e-27
    SPR.vertline.P09076.vertline.HME3_APIME HOMEOBOX PROTEIN E30 (FRAGMENT).
     5.4e-27
    SPR.vertline.P31533.vertline.HME3_BRARE HOMEOBOX PROTEIN ENGRAILED-3 (ZF-E
     . . . 2.0e-26
    SPR.vertline.P31536.vertline.HMEB_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE B .
     . . 8.8e-26


This time we add HMEN_LAMPL to our cluster, and we start the next BLASTP search with this sequence, yielding the following result (excerpt): 

    Sequences found:
     Probability:
    SPR.vertline.P31534.vertline.HMEN_LAMPL HOMEOBOX PROTEIN ENGRAILED-LIKE (E
     . . . 5.7e-37
    SPR.vertline.P31535.vertline.HMEA_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE A .
     . . 5.0e-31
    SPR.vertline.P31538.vertline.HMEB_XENLA HOMEOBOX PROTEIN ENGRAILED-1B (EN-
     . . . 1.4e-28
    SPR.vertline.P31537.vertline.HMEA_XENLA HOMEOBOX PROTEIN ENGRAILED-1A (EN-
     . . . 2.9e-28
    SPR.vertline.P23397.vertline.HMEN_HELTR HOMEOBOX PROTEIN HT-EN (FRAGMENT).
     1.2e-27
    SPR.vertline.P31536.vertline.HMEB_MYXGL HOMEOBOX PROTEIN ENGRAILED-LIKE B .
     . . 1.4e-27
    SPR.vertline.Q04896.vertline.HME1_BRARE HOMEOBOX PROTEIN ENGRAILED-1.
     1.5e-27
    SPR.vertline.P09015.vertline.HME2_BRARE HOMEOBOX PROTEIN ENGRAILED-2 (ZF-E
     . . . 6.9e-27
    SPR.vertline.Q05925.vertline.HME1_HUMAN HOMEOBOX PROTEIN ENGRAILED-1 (HU-E
     . . . 1.5e-26
    SPR.vertline.P09065.vertline.HME1_MOUSE HOMEOBOX PROTEIN ENGRAILED-1 (MO-E
     . . . 1.6e-26
    SPR.vertline.P09075.vertline.HME6_APIME HOMEOBOX PROTEIN E60 (FRAGMENT).
     1.9e-26
    SPR.vertline.Q05916.vertline.HME1_CHICK HOMEOBOX PROTEIN ENGRAILED-1 (GG-E
     . . . 4.5e-26


Above the threshold, we do not find any sequences that would not already be contained in our cluster, so the SYSTERS search for this inquiry sequence is now concluded, and the cluster contains the following 26 sequences: 

    HME1_HUMAN,       HME1_MOUSE,       HME1_CHICK,     HME2_CHICK,
     HME2_HUMAn,
    HME2_MOUSE,       HME2_BRARE,       HMEB_XENLA,     HMEC_XENLA,
     HMED_XENLA,
    HME3_BRARE,       HME1_BRARE,       HMEN_DROVI,     HMIN_DROME,
     HMEN_BOMMO,
    HMEN_DROME,       HMIN_BOMMO,       HMEN_TRIGR,     HMEN_ARTSF,
     HME3_APIME,
    HME6_APIME,       HMEN_SCHAM,       HMEN_HELTR,     HMEA_XENLA,
     HMEA_MYXGL,
    HMEN_LAMPL.


If this procedure is performed for all 28 sequences annotated as homeobox engrailed in the Swissprot database, this yields 28 clusters at first. The clusters thus found are plotted in the following table against the sequences, where the columns represent the clusters belonging to the inquiry sequence listed at the head of the table and the line indicate the clusters in which the sequence listed at the left is contained (marked with an X). In this case, there are seven clusters having 27 sequences each, five clusters having 26 sequences each, etc. 

        Cluster (inquiry sequence)
    Sequence           1 (HME2_BRARE)     2 (HME2_MOUSE)     3 (HMEB_XENLA)
     4 (HMEC_XENLA)     5 (HMED_XENLA)     6 (HMEN_LAMPL)     7 (HMEA_MYXGL)
      8 (HMEA_XENLA)
    HME2_BRARE                X                 X                  X
           X                  X                  X                  X
            X
    HME2_MOUSE                X                 X                  X
           X                  X                  X                  X
            X
    HMEB_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HMEC_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HMED_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_LAMPL                X                 X                  X
           X                  X                  X                  X
    HMEA_MYXGL                X                 X                  X
           X                  X                  X                  X
            X
    HMEA_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HM16_CAEEL                X                 X                  X
           X                  X                  X                  X
            X
    HME1_MOUSE                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_DROME                X                 X                  X
           X                  X                  X                  X
            X
    HMIN_DROME                X                 X                  X
           X                  X                  X                  X
            X
    HME6_APIME                X                 X                  X
           X                  X                  X                  X
            X
    HME3_APIME                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_DROVI                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_TRIGR                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_SCHAM                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_HELTR                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_BOMMO                X                 X                  X
           X                  X                  X                  X
            X
    HME3_BRARE                X                 X                  X
           X                  X                  X                  X
            X
    HME1_BRARE                X                 X                  X
           X                  X                  X                  X
            X
    HMIN_BOMMO                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_ARTSF                X                 X                  X
           X                  X                  X                  X
            X
    HME2_HUMAN                X                 X                  X
           X                  X                  X                  X
            X
    HME1_CHICK                X                 X                  X
           X                  X                  X                  X
            X
    HME2_CHICK                X                 X                  X
           X                  X                  X                  X
            X
    HME1_HUMAN                X                 X                  X
           X                  X                  X                  X
            X
    HMEB_MYXGL
        Cluster (inquiry sequence)
    Sequence           9 (HM16_CAEEL)    10 (HME1_MOUSE)    11 (HMEN_DROME)
     12 (HMIN_DROME)    13 (HME6_APIME)    14 (HME3_APIME)    15 (HMEN_DROVI)
      16 (HMEN_TRIGR)
    HME2_BRARE                X                 X                  X
           X                  X                  X                  X
            X
    HME2_MOUSE                X                 X                  X
           X                  X                  X                  X
            X
    HMEB_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HMEC_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HMED_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_LAMPL
    HMEA_MYXGL                                  X
    HMEA_XENLA                X                 X                  X
           X                  X                  X                  X
            X
    HM16_CAEEL                X                 X                  X
           X                  X                  X                  X
            X
    HME1_MOUSE                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_DROME                X                 X                  X
           X                  X                  X                  X
            X
    HMIN_DROME                X                 X                  X
           X                  X                  X                  X
            X
    HME6_APIME                X                 X                  X
           X                  X                  X                  X
            X
    HME3_APIME                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_DROVI                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_TRIGR                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_SCHAM                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_HELTR                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_BOMMO                X                 X                  X
           X                  X                  X                  X
            X
    HME3_BRARE                X                 X                  X
           X                  X                  X                  X
            X
    HME1_BRARE                X                 X                  X
           X                  X                  X                  X
            X
    HMIN_BOMMO                X                 X                  X
           X                  X                  X                  X
            X
    HMEN_ARTSF                X                 X                  X
           X                  X                  X                  X
            X
    HME2_HUMAN                X                 X                  X
           X                  X                  X                  X
            X
    HME1_CHICK                X                 X                  X
           X                  X                  X                  X
            X
    HME2_CHICK                X                 X                  X
           X                  X                  X                  X
            X
    HME1_HUMAN                X                 X                  X
           X                  X                  X                  X
            X
    HMEB_MYXGL
        Cluster (inquiry sequence)
    Sequence            17 (HMEN_SCHAM)      18 (HMEN_HELTR)        19
     (HMEN_BOMMO)      20 (HME3_BRARE)        21 (HME1_BRARE)      22
     (HMIN_BOMMO)      (23 (HMEN_ARTSF)
    HME2_BRARE                 X                    X                     X
                     X                     X                     X
           X
    HME2_MOUSE                 X                    X                     X
                     X                     X                     X
           X
    HMEB_XENLA                 X                    X                     X
                     X                     X                     X
           X
    HMEC_XENLA                 X                    X                     X
                     X                     X                     X
           X
    HMED_XENLA                 X                    X                     X
                     X                     X                     X
           X
    HMEN_LAMPL
    HMEA_MYXGL
    HMEA_XENLA                 X                    X                     X
                     X                     X
    HM16_CAEEL                 X                    X                     X
                     X                     X                     X
    HME1_MOUSE                 X                    X                     X
                     X                     X                     X
           X
    HMEN_DROME                 X                    X                     X
                     X                     X                     X
           X
    HMIN_DROME                 X                    X                     X
                     X                     X                     X
           X
    HME6_APIME                 X                    X                     X
                     X                     X                     X
           X
    HME3_APIME                 X                    X                     X
                     X                     X                     X
           X
    HMEN_DROVI                 X                    X                     X
                     X                     X                     X
           X
    HMEN_TRIGR                 X                    X                     X
                     X                     X                     X
           X
    HMEN_SCHAM                 X                    X                     X
                     X                     X                     X
           X
    HMEN_HELTR                 X                    X                     X
                     X                     X                     X
           X
    HMEN_BOMMO                 X                    X                     X
                     X                     X                     X
           X
    HME3_BRARE                 X                    X                     X
                     X                     X                     X
           X
    HME1_BRARE                 X                    X                     X
                     X                     X                     X
           X
    HMIN_BOMMO                 X                    X                     X
                     X                     X                     X
           X
    HMEN_ARTSF                 X                    X                     X
                     X                     X                     X
           X
    HME2_HUMAN                 X                    X                     X
                     X                     X                     X
           X
    HME1_CHICK                 X                    X                     X
                     X                     X                     X
           X
    HME2_CHICK                 X                    X                     X
                     X                     X                     X
           X
    HME1_HUMAN                 X                    X                     X
                     X                     X                     X
           X
    HMEB_MYXGL
                                   Cluster (inquiry sequence)
                                                           Sequence
     24 (HME2_HUMAN)    25 (HME1_CHICK)    26 (HME2_CHICK)    27 (HME1_HUMAN)
      28 (HMEB_MYXGL)
                                                           HME2_BRARE
          X                  X                  X                  X
                                                           HME2_MOUSE



          X                  X                  X                  X
                                                           HMEB_XENLA
          X                  X                  X                  X
                                                           HMEC_XENLA
          X                  X                  X                  X
                                                           HMED_XENLA
          X                  X                  X                  X
                                                           HMEN_LAMPL
          X                  X                  X                  X
                                                           HMEA_MYXGL
          X                  X                  X                  X
                                                           HMEA_XENLA
          X                  X                  X                  X
                                                           HM16_CAEEL
                                                           HME1_MOUSE
          X                  X                  X                  X
                                                           HMEN_DROME
          X                  X                  X                  X
                                                           HMIN_DROME
          X                  X                  X                  X
                                                           HME6_APIME
          X                  X                  X                  X
                                                           HME3_APIME
          X                  X                  X                  X
                                                           HMEN_DROVI
          X                  X                  X                  X
                                                           HMEN_TRIGR
          X                  X                  X                  X
                                                           HMEN_SCHAM
          X                  X                  X                  X
                                                           HMEN_HELTR
          X                  X                  X                  X
                                                           HMEN_BOMMO
          X                  X                  X                  X
                                                           HME3_BRARE
          X                  X                  X                  X
                                                           HME1_BRARE
          X                  X                  X                  X
                                                           HMIN_BOMMO
          X                  X                  X                  X
                                                           HMEN_ARTSF
          X                  X                  X                  X
                                                           HME2_HUMAN
          X                  X                  X                  X
                                                           HME1_CHICK
          X                  X                  X                  X
                                                           HME2_CHICK
          X                  X                  X                  X
                                                           HME1_HUMAN
          X                  X                  X                  X
                                                           HMEB_MYXGL
           X


After removing identical clusters and solving for inclusions, the homeobox engrailed proteins are distributed among two clusters--one with 27 sequences and the other with only the HMEA_MYXGL sequence.

«Previous Next»
Method for enforcing integrity constraints in a database table using an index System and method for enhanced performance of a relational database management system through the use of application-specific memory-resident data