Library preparation and use thereof for sequencing based error correction and/or variant identification

1. A method for correcting sequencing derived errors in sequence reads, the method comprising:

(a) obtaining a test sample comprising a plurality of cell-free DNA (cfDNA) molecules comprising a plurality of double-stranded DNA (dsDNA) fragments;

(b) preparing a sequencing library, wherein preparing the sequencing library comprises:

(i) providing a set of loop-shaped DNA adapters, wherein the set of loop-shaped DNA adapters comprises:

a plurality of first loop-shaped DNA adapters where each of the first loop-shaped DNA adapters comprises a single DNA molecule comprising:

(A) two complementary regions that hybridize with one another and leave an unpaired loop at the end of the DNA molecule;

(B) an endonuclease restriction site in the unpaired loop; and

(C) a single first unique molecular identifier (UMI);

a plurality of second loop-shaped DNA adapters where each of the second loop-shaped DNA adapters comprises a single DNA molecule comprising:

(A) two complementary regions that hybridize with one another and leave an unpaired loop at the end of the DNA molecule; and

(B) a single second unique molecular identifier (UMI);

(ii) ligating the plurality of first loop-shaped DNA adapters to a first end of the dsDNA fragments and ligating the plurality second loop-shaped DNA adapters to a second end of the dsDNA fragments to generate a plurality of circular-shaped constructs, wherein each circular-shaped construct comprises a first loop-shaped DNA adapter ligated to a first end of the dsDNA fragment and a second loop-shaped DNA adapter ligated to a second end of the dsDNA fragment; and

(iii) after step (ii), cleaving the plurality of first loop-shaped DNA adapters with an endonuclease to produce a plurality of linear single-stranded DNA (ssDNA) molecules, wherein said linear ssDNA molecules comprise a forward strand and a reverse complement strand;

(c) sequencing a plurality of the linear ssDNA molecules in the sequencing library to generate a plurality of sequence reads;

(d) grouping the plurality of sequence reads into a plurality of families based on the first UMI and the second UMI, such that one or more unique nucleic acid sequence fragments originating from the same test sample contains the first UMI and the second UMI;

(i) including in the plurality of families each of the sequence reads that comprises both the first UMI and the second UMI, and

(ii) excluding from the plurality of families sequence reads that only comprise the first UMI on both ends of a dsDNA fragment or the second UMI on both ends of the dsDNA fragment;

(e) comparing the forward strand and the reverse complement strand of each of the sequence reads within each family of the plurality of families to generate a consensus sequence for each family, wherein the consensus sequence comprises a sequence of nucleotide bases, and wherein each nucleotide base is identified at a given position in the consensus sequence when a specific nucleotide base is present at the position in at least 70% of the sequence reads of family members within each family of the plurality of families;

(f) aligning the consensus sequence for each family of the plurality of families to a reference sequence; and

(g) identifying a consensus sequence as comprising a sequencing-derived error in the sequence reads if the consensus sequence differs from the reference sequence at one or more nucleotide base positions.

2. The method according to claim 1 , wherein the consensus sequence for each family comprises a sequence of nucleotide bases, wherein each nucleotide base is identified at a given position in the consensus sequence when a specific nucleotide base is present in at least 80% of the sequence reads of family members within each family of the plurality of families.

3. The method according to claim 1 , wherein the consensus sequence comprises a sequence of nucleotide bases, wherein each nucleotide base is identified at a given position in the consensus sequence when a specific nucleotide base is present in at least 90% of the sequence reads of family members within each family of the plurality of families.

4. The method according to claim 1 , wherein the consensus sequence comprises a sequence of nucleotide bases, wherein each nucleotide base is identified at a given position in the consensus sequence when a specific nucleotide base is present in at least 95% of the sequence reads of family members within each family of the plurality of families.

5. The method according to claim 1 , wherein the step of sequencing a plurality of the linear ssDNA molecules in the sequencing library comprises sequencing by a next-generation sequencing (NGS) procedure.

6. The method according to claim 1 , wherein the step of sequencing a plurality of the linear ssDNA molecules in the sequencing library comprises a sequencing-by-synthesis procedure.

7. The method according to claim 1 , wherein the step of sequencing a plurality of the linear ssDNA molecules in the sequencing library comprises a paired-end sequencing procedure.

8. The method of claim 1 , wherein the step of sequencing a plurality of the linear ssDNA molecules in the sequencing library comprises a single molecule sequencing procedure.

9. The method of claim 5 , wherein the NGS procedure comprises single-molecule real-time sequencing.