Nucleic acid sequence assembly

Abstract

Disclosed herein are compositions, systems and methods related to sequence assembly, such as nucleic acid sequence assembly of single reads and contigs into larger contigs and scaffolds through the use of read pair sequence information, such as read pair information indicative of nucleic acid sequence phase or physical linkage.

What is claimed is: 
     
         1 . A method for nucleic acid sequence assembly, comprising:
 (a) obtaining purified DNA;   (b) binding the purified DNA with a DNA binding agent to form DNA/chromatin complexes;   (c) incubating the DNA-chromatin complexes with restriction enzymes to leave sticky ends;   (d) performing ligation to join ends of DNA;   (e) sequencing across ligated DNA junctions to generate paired end reads;   (f) obtaining standard paired-end read distance frequency data; and   (g) scaffolding the grouped contig sequences such that read pair distance frequency data for read pairs that map to separate contigs approximates the standard paired-end read distance frequency data.   
     
     
         2 . The method of  claim 1 , wherein read pair distance frequency data for read pairs that map to separate contigs more closely approximates the paired-end read distance frequency data when read pair distance likelihood is maximized. 
     
     
         3 . A method for scaffolding contigs of nucleic acid sequence information obtained from a biological sample, said method comprising:
 obtaining a set of contig sequences having an initial configuration, wherein the contig sequences are obtained by extracting DNA from a biological material and sequencing the biological material;   obtaining a set of paired end reads, wherein the set of paired-end reads is obtained by digesting sample DNA to generate internal double strand breaks within the nucleic acid, allowing the double strand breaks to re-ligate randomly to form a plurality of re-ligation junctions, and sequencing across the plurality of re-ligation junctions;   obtaining standard paired-end read distance frequency data;   grouping contig pairs sharing sequence that coexists in at least one paired end read;   and scaffolding the grouped contig sequences such that read pair distance frequency data for read pairs that map to separate contigs more closely approximates the standard paired-end read distance frequency data relative to the read pair frequency data of the contig sequences in the initial configuration.   
     
     
         4 . The method of  claim 3 , wherein the sample DNA is crosslinked to at least one DNA binding agent. 
     
     
         5 . The method of  claim 3 , wherein the sample DNA comprises isolated naked DNA. 
     
     
         6 . The method of  claim 5 , wherein the isolated naked DNA is reassembled into reconstituted chromatin. 
     
     
         7 . The method of  claim 6 , wherein the reconstituted chromatin is crosslinked. 
     
     
         8 . The method of  claim 6 , wherein the reconstituted chromatin comprises a DNA binding protein. 
     
     
         9 . The method of  claim 6 , wherein the reconstituted chromatin comprises a nanoparticle. 
     
     
         10 . The method of  claim 3 , wherein the set of paired-end reads are obtained by digesting sample DNA to generate internal double strand breaks within the nucleic acid, allowing the double strand breaks to re-ligate randomly to form a plurality of re-ligation junctions, and sequencing across the plurality of re-ligation junctions. 
     
     
         11 . The method of  claim 3 , wherein standard paired-end read distance frequency data is obtained from paired-end reads where both reads map to a common contig. 
     
     
         12 . The method of  claim 3 , wherein standard paired-end read distance frequency data is obtained from previously generated curves. 
     
     
         13 . The method of  claim 3 , wherein said scaffolding comprises selecting a first set of putative adjacent contigs of said clustered contigs, determining a minimal distance order of said first set of putative adjacent contigs that reduces an aggregate measure of the read-pair distances for said read pairs, and scaffolding said first set of putative adjacent contigs so as to reduce said aggregate measure of the read-pair distance. 
     
     
         14 . The method of  claim 13 , wherein said determining a minimal distance order comprises comparing the expected read-pair distance for at least one read pair that comprises reads mapping to two contigs of said set for all possible contig configurations. 
     
     
         15 . The method of  claim 14 , comprising selecting the contig orientation that corresponds to the maximum likelihood read pair distance distribution. 
     
     
         16 . The method of  claim 3 , wherein the nucleic acid sequence information is derived from a genome. 
     
     
         17 . The method of  claim 3 , wherein the nucleic acid sequence information is derived from a heterogeneous sample comprising a plurality of genomes. 
     
     
         18 . A method for scaffolding contigs of nucleic acid sequence information comprising:
 obtaining a set of contig sequences having an initial configuration;   obtaining a set of paired end reads;   obtaining standard paired-end read distance frequency data;   grouping contig pairs sharing sequence that coexists in at least one paired end read;   and scaffolding the grouped contig sequences such that read pair distance frequency data for read pairs that map to separate contigs more closely approximates the standard paired-end read distance frequency data relative to the read pair frequency data of the contig sequences in the initial configuration.   
     
     
         19 . The method of  claim 18 , wherein the scaffolding comprises at least one of ordering the set of contigs, orienting the set of contigs, merging at least two contigs end to end, inserting one contig into a second contig, and cleaving a contig into at least two constituent contigs. 
     
     
         20 . The method of  claim 18 , wherein read pair distance frequency data for read pairs that map to separate contigs more closely approximates the paired-end read distance frequency data when read pair distance likelihood increases. 
     
     
         21 . The method of  claim 20 , wherein read-pair distance likelihood is maximized. 
     
     
         22 . The method of  claim 18 , wherein read pair distance frequency data for read pairs that map to separate contigs more closely approximates the paired-end read distance frequency data when a statistical measure of difference between the read pair distance frequency data and the standard paired-end read distance frequency data decreases. 
     
     
         23 . The method of  claim 22 , wherein the statistical measure of distance between the read pair distance frequency data and the standard paired-end read distance frequency data comprises at least one of ANOVA, a t-test, and a X-squared test. 
     
     
         24 . The method of  claim 23 , wherein read pair distance for read pairs that map to separate contigs more closely matches the paired-end read distance frequency data when deviation of read pair distance distribution among ordered contigs obtained as compared to standard paired-end read distance frequency decreases. 
     
     
         25 . The method of  claim 24 , wherein deviation of read pair distance distribution among ordered contigs obtained as compared to standard paired-end read distance frequency is minimized. 
     
     
         26 . A method of assembling contig sequence information into at least one scaffold, comprising
 obtaining sequence information corresponding to a plurality of contigs,   obtaining paired-end read information from a nucleic acid sample represented by the plurality of contigs,   and configuring the plurality of contigs such that deviation of a read pair distance parameter from a predicted read pair distance data set is minimized,   wherein the configuring occurs in less than 8 hours.   
     
     
         27 . The method of  claim 26 , wherein the predicted read pair distance data set comprises a read pair distance likelihood curve. 
     
     
         28 . The method of  claim 26 , wherein the read pair distance parameter is maximum distance likelihood relative to a read pair distance likelihood curve. 
     
     
         29 . The method of  claim 26 , wherein the read pair distance parameter is minimum variation relative to a read pair distance likelihood curve. 
     
     
         30 . The method of  claim 26 , wherein 70% of the set of contig sequences are ordered and oriented so as to match the relative order and orientation of their sequences in the nucleic acid sample in no more than 8 hours.