North Carolina State University

Complete Genomics, Inc.

Illumina Corporation

Sicel Technologies, Inc.

Other

Guardant Health, Inc.

Snc Holdings Corp.

Bahram Ghaffarzadeh Kermani

Charles W Scarantino

H Troy Nagle

Radoje Drmanac

Jeffrey S Mueller

Lester C Hall

Helmy A Eltoukhy

Brock A Peters

Mark S Chee

Amirali Talasaz

Jason Oliver Fiering

Steven Mark Barnard

Hongji Ren

AmirAli Talasaz

Todd Alan Dickinson

Juergen Haas

Stefan Ufer

Anthony W Czarnik

Shawn O Meade

Shawn Christopher Baker

Theo Kotseroglou

Timothy K McDaniel

Nels A Olson

Chang-Soo Kim

Karen W Shannon

James Bierle

Oleg Alferov

Nnamdi Ihuegbu

Troy H Nagle

Mohammad Reza Mokhtari

Leslie C Hall

Jeffrey Mueller

Display screen or portion thereof with graphical user interface

Systems and methods are disclosed to detect single-nucleotide variations (SNVs) from somatic sources in a cell-free biological sample of a subject by generating training data with class labels; in computer memory, generating a machine learning unit comprising one output for each of adenine (A), cytosine (C), guanine (G), and thymine (T) calls; training the machine learning unit; and applying the machine learning unit to detect the SNVs from somatic sources in the cell-free biological sample of the subject, wherein the cell-free biological sample comprises a mixture of nucleic acid molecules from somatic and germline sources.

Machine learning for somatic single nucleotide variant detection in cell-free tumor nucleic acid sequencing applications

Techniques for DNA-based storage of electronic data are described herein. In an example embodiment, a plurality of files is stored in deoxyribonucleic acid (DNA)-based storage. The plurality of files is encoded in a set of DNA oligos, where a DNA synthesizer system synthesizes first DNA oligos that encode first type of segments from the plurality of files and second DNA oligos that encode second type of segments from the plurality of files, and where the first DNA oligos are synthesized in excess compared to the second DNA oligos.

Accurate and efficient DNA-based storage of electronic data

Techniques perform de novo assembly. The assembly can use labels that indicate origins of the nucleic acid molecules. For example, a representative set of labels identified from initial reads that overlap with a seed can be used. Mate pair information can be used. A sequence read that aligns to an end of a contig can lead to using the other sequence read of a mate pair, and the other sequence read can be used to determine which branch to use to extend, e.g., in an external cloud or helper contig. A kmer index can include labels indicating an origin of each of the nucleic acid molecules that include each kmer, memory addresses of the reads that correspond to each kmer in the index, and a position in each of the mate pairs that includes the kmer. Haploid seeds can also be determined using polymorphic loci identified in a population.

Long fragment de novo assembly using short reads

Techniques for biochemcally-enabled security/authentication mechanisms are described herein. In an example embodiment, a security system receives a biological sample from a key. The biological sample includes a set of deoxyribonucleic acid (DNA) oligos that represent a code assigned to the key. The set of DNA oligos is sequenced to obtain a set of read sequences. The set of read sequences is then filtered to identify a set of filtered sequences. The set of filtered sequences is matched to sets of expected sequences, where the sets of expected sequences are assigned to respective keys issued for the security system. Access to a resource is then granted or denied based on whether the set of filtered sequences matches with any set from the sets of expected sequences.

Method and apparatus for DNA-based authentication system

Long fragment read techniques can be used to identify deletions and resolve base calls by utilizing shared labels (e.g., shared aliquots) of a read with any reads corresponding to heterozygous loci (hets) of a haplotype. For example, the linking of a locus to a haplotype of multiple hets can increase the reads available at the locus for determining a base call for a particular haplotype. For a hemizygous deletion, a region can be linked to one or more hets, and the labels for a particular haplotype can be used to identify which reads in the region correspond to which haplotype. In this manner, since the reads for a particular haplotype can be identified, a hemizygous deletion can be determined. Further, a phasing rate of pulses can be used to identify large deletions. A deletion can be identified with the phasing rate is sufficiently low, and other criteria can be used.

Phasing and linking processes to identify variations in a genome

Techniques for DNA-based storage of electronic data are described herein. In an example embodiment, a file system is stored in deoxyribonucleic acid (DNA)-based storage. The file system is encoded in a set of DNA oligos, where a DNA synthesizer system synthesizes first DNA oligos that encode metadata of the file system and second DNA oligos that encode the contents of files in the file system.

Methods, systems, and apparatuses are provided for creating and using a machine-leaning model to call a base at a position of a nucleic acid based on intensity values measured during a production sequencing run. The model can be trained using training data from training sequencing runs performed earlier. The model is trained using intensity values and assumed sequences that are determined as the correct output. The training data can be filtered to improve accuracy. The training data can be selected in a specific manner to be representative of the type of organism to be sequenced. The model can be trained to use intensity signals from multiple cycles and from neighboring nucleic acids to improve accuracy in the base calls.

Basecaller for DNA sequencing using machine learning

Haplotypes of one or more portions of a chromosome of an organism from sequencing information of DNA or RNA fragments can be determined. Heterozygous loci (hets) can be used to determine haplotypes. One allele on a first het can be connected (likely to be on the same haplotype) to an allele on a second het, thereby defining a particular orientation between the hets. Haplotypes can be assembled through these connections. Errors can be identified through redundant connection information, particularly using a confidence value (strength) for a particular connection. The connections among a set of hets can be analyzed to determine likely haplotypes for that set, e.g., an optimal tree of a graph containing the hets. Furthermore, haplotypes of different contiguous sections (contig) of the chromosome can be matched to a particular chromosome copy (e.g., to a particular parental copy). Thus, the phase of an entire chromosome can be determined.

Phasing of heterozygous loci to determine genomic haplotypes

Oncology monitoring systems include: (a) a first plurality of radiation sensors configured to reside proximate a target tumor treatment site of a patient, the sensors being configured to provide radiation data associated with the tumor treatment site; and (b) a first portable receiver in communication with the plurality of sensors. The receiver is configured to obtain radiation data from the sensors at a plurality of different times. The receiver is in communication with a local and/or remote computer that tracks variation in the radiation data to provide dynamic tumor site information.

Methods and systems for monitoring patients undergoing treatment for cancer

The present invention is directed to logic for analysis of nucleic acid sequence data that employs algorithms that lead to a substantial improvement in sequence accuracy and that can be used to phase sequence variations, e.g., in connection with the use of the long fragment read (LFR) process.

Processing and analysis of complex nucleic acid sequence data

Various short reads can be grouped and identified as coming from a same long DNA fragment (e.g., by using wells with a relatively low-concentration of DNA). A histogram of the genomic coverage of a group of short reads can provide the edges of the corresponding long fragment (pulse). The knowledge of these pulses can provide an ability to determine the haploid genome and to identify structural variations.

Identification of DNA fragments and structural variations

Sequencing information is used to correlate alleles at certain locations to alleles at other locations. The statistical information from the reads of fragments in a sample can be used to determine the phasing of haplotypes and to correct or confirm based calls at the locations. In one example, a confidence value (strength score) is determined for a particular hypothesis, which can include whether two alleles are on a same haplotype at two particular loci, as well as what the alleles are on another haplotype (e.g. for a diploid organism). The strength can include a positive contribution from data that is consistent with the hypothesis and a negative contribution from data is that inconsistent with the hypothesis, where both values can be used in a formula to determine the strength.

Analyzing genome sequencing information to determine likelihood of co-segregating alleles on haplotypes

Data extracted from fluorosphore responses of fluorophore labeled bases in genetic material used in sequencing of unknown fragments from a defined set of for example a model system are converted into a class of block codes that are then employed in a computer-based process to compare and correct preliminary calls of calls of the categorically known genetic material. In a specific embodiment, the Reed-Solomon codes are employed to identify one or more errors as may occur in a finite block of codes. The methodology is also useful to identify elements of a real system containing known elements in the form of a tag. Reed-Solomon sensors may be employed with and in addition to other types of genome sensors.

Method and apparatus for quantification of DNA sequencing quality and construction of a characterizable model system using Reed-Solomon codes

Growing community of inventors

Los Altos, CA, United States of America

Bahram Ghaffarzadeh Kermani