Selected article for: "index reference and reference list"

Author: Marie Hoffmann; Michael T. Monaghan; Knut Reinert
Title: PriSeT: Efficient De Novo Primer Discovery
  • Document date: 2020_4_7
  • ID: 3b3hv53b_46
    Snippet: After the FM index computation, the original reference sequences are only needed once to lookup and encode frequent k-mers (see Filter Step). However, when combining k-mer candidates to form pairs, we need location informationbecause k-mers need to refer to the same reference and have to be in an offset range of [Ï„ min : Ï„ max ] nt. We use two data structures per reference -a list to store the set of TKMerIDs in order of occurrence and a compac.....
    Document: After the FM index computation, the original reference sequences are only needed once to lookup and encode frequent k-mers (see Filter Step). However, when combining k-mer candidates to form pairs, we need location informationbecause k-mers need to refer to the same reference and have to be in an offset range of [Ï„ min : Ï„ max ] nt. We use two data structures per reference -a list to store the set of TKMerIDs in order of occurrence and a compact data structure in form of a bit vector B in the length of the last k-mer occurrence. A set bit indicates the presence of a TKMerID. I.e., the i-th set bit in B corresponds to the i-th TKMerID in the KMerIDs list. The compact data structure is augmented with rank 1 and select 1 support, which have O(1) runtime for queries 10 . Figure 3 illustrates the k-mer combination search over the transformed reference.

    Search related documents:
    Co phrase search for related documents
    • compact data and data structure: 1
    • compact data structure and data structure: 1
    • data structure and Filter Step: 1
    • FM index computation and index computation: 1, 2, 3, 4