DE19950050C2 - Method for the functional assignment of unclassified DNA sequences - Google Patents

Description

The present invention relates to a method for the functional assignment of unclassified DNA sequences using an unclassified sequence simple steps can be assigned to known reference sequence (aligned).

Since the commercial availability of PCR technology, the functional Assignment of the DNA information made available in this way fundamental problem of biotechnology.

In conventional processes, therefore, is either about functional peculiarities or direct comparison of the unclassified sequence with sequences with known ones Matched property.

The present application provides a method for the functional assignment of an unclassified DNA sequence, which comprises the following steps:

• a) Compare the unclassified sequence (A) with reference sequences (B ₁ -B _n ) below
• b) creating gap patterns (C ₁ -C _m ) for the reference sequences (B ₁ -B _n ) and a consensus sequence (D) for the sequence (A),
• c) splitting the gap patterns (C ₁ -C _m ) into short consensus sequences (E ₁ -E _m ) and gap information (F ₁ -F _m ),
• d) position-by-point comparison of the short consensus sequences (E ₁ -E _m ) with iterating offset with the unclassified sequence (A) while determining the short consensus sequence with the highest agreement (E _max ),
• e) inserting the corresponding one of the consensus sequence (E _max) gap information (F _max) in the sequence (A) to create a Aligned sequence (G).

The reference sequences (B ₁ -B _n ) on which the consensus sequence (D) is based should have a degree of agreement among themselves which is greater than 60 and less than 80-90%. According to the present application, the information collected in step (b) (gap pattern (C ₁ -C _m ) and consensus sequence (E ₁ -E _m ) can be buffered and used directly for later comparisons.

When creating the gap patterns, the same gap patterns are eliminated, so that for E ₁ -E _{m the} following applies: m ≦ n.

The critical step in the method according to the invention is the positional comparison of the short consensus sequences (E ₁ -E _m ) to achieve the highest possible degree of agreement. For a meaningful classification it is necessary that this degree of agreement is as large as possible greater than 60%, preferably greater than 80%.

The method according to the invention can also be carried out in several cycles, and after finding the aligned sequence (G), a best reference sequence (B _max ) is found, which is used to find a new set of reference sequences (H ₁ -H _n ) belonging to the family belongs to the reference sequence with the highest degree of agreement (B _max ) of the first cycle.

The method offers the advantage of the very fast generation of multiples Sequence alignments so that they can be used very quickly for further sequence processing such as B. sequence annotation or for sequence comparisons are available. Especially sequence comparisons in correctly calculated multiples Sequence alignments are very fast since the sequences in the alignment for the Sequence comparison no longer need to be shifted against each other, but Position by position can be compared directly.

The invention is explained in more detail with reference to the following figures:

Fig. 1 Reading the sequences First, the sequences to be processed and the reference sequences are read.

Fig. 2 The gap patterns are extracted from the reference sequences and a consensus sequence is determined from the sequence information.

Fig. 3a A list of consensus sequences is then created, from which a particular gap pattern is removed in each case.

Fig. 3B shows a specific example of how to generate consensus sequence from Shorted Konsensi.

FIG. 4a An optimal set of aligner parameters is determined by comparing each of the sequences in "Shorted Consensus" sequences with the new sequence.

Fig. 4b shows a specific example for the positional comparison.

Fig. 5a The new sequence is aligned by inserting the gaps and shifting the offset in accordance with the parameter set.

Fig. 5b Experimental example of inserting the gaps and offsets of the offset.

Claims (4)

1. A method for functionally mapping an unclassified DNA sequence comprising:

• a) Compare the unclassified sequence (A) with reference sequences (B ₁ -B _i ) below
• b) creating gap patterns (C ₁ -C _m ) for the reference sequences (B ₁ -B _n ) and a consensus sequence (D) for the sequence (A),
• c) splitting the gap patterns (C ₁ -C _m ) into short consensus sequences (E ₁ -E _m ) and gap information (F ₁ -F _m ),
• d) position-by-point comparison of the short consensus sequences (E ₁ -E _m ) with iterating offset with the sequence (A) while determining the short consensus sequence with the highest agreement (E _max ),
• e) inserting the corresponding one of the consensus sequence (E _max) gap information (F _max) in the sequence (A) to create a Aligned sequence (G).

2. The method of claim 1, wherein in the positional comparison Degree of agreement of ≧ 60%, preferably ≧ 80%, is required. 3. The method according to claim 1 or 2, wherein several cycles of steps (a) to (e) respectively. 4. The method according to claim 1, wherein the gap pattern (C ₁ -C _m ) and consensus sequence (D) created in step (b) are used for later comparisons.