GRAIL, LLC

Compositions and methods are provided for amplifying nucleic acids, including cell free nucleic acid fragments, in preparation for sequencing. Methods are provided for making circularized nucleic acid templates having the structure [T]-[PS1]-[L]-[PS2] or [PS1]-[L]-[PS2]-[T′], where (a) T is a target nucleic acid and T′ is a complement to a target nucleic acid; (b) each of PS1 and PS2 is a nucleic acid primer site; (c) L is a linker having a primer extension reaction terminating organic molecule; and the structure is circularized by binding a 5′ end thereof to a 3′ end thereof. Target sequences in the circularized templates are amplified by binding to PS1 a primer complimentary to PS1 and binding to PS2 a primer complimentary to PS2 and copying the target sequences by a primer extension reaction. Advantages include a reduction in ligation steps, which can result in fewer clean up steps and improved library conversion efficiency.

The present disclosure relates to an improved method for analyzing sequencing data to detect cross-sample contamination in a test sample. Determining cross-contamination in a test sample can be informative for determining that the test sample will be less likely to correctly identify the presence of cancer in the subject. Pre-determined single nucleotide polymorphisms selected from: an allele present in a select database or a genotyping SNP associated with a sample type are used to identify. A sample is determined to be contaminated using the determined contamination probabilities of the one or more pre-determined SNPs.

A system generates a cancer detection panel. The system is configured to generate an assay having a minimized size and number of genomic regions while still detecting the presence of cancer at or above a specific performance threshold. To select the genomic regions for the panel, the system employs a classification model. The classification model receives a set of genomic regions that may be associated with disease presence. The model then determines a sensitivity score for each genomic region and ranks the regions according to their score. The sensitivity score is based on a likelihood that variations in the genomic region are indicative of cancer. The model then selects genomic regions for the panel based on their rank. The model only selects as many genomic indicators as are needed for desired detection performance. The genomic regions can be associated with solid or liquid cancers, viral regions, or cancer hotspots.

Methods and systems for segmenting sequencing regions obtained from a sample interval are disclosed. sample contamination detection are disclosed. In particular, an analytics system accesses test sequences from a sample. The test sequences each include a sequencing region which, in aggregate, form an aggregate sequencing region. The analytics system segments sequencing regions from the aggregate sequencing region into sequencing subregions. Several methods of segmenting sequencing regions into sequencing subregions are disclosed: (1) maximizing cancer vs. non-cancer methylation beta differences, (2) minimizing cancer vs. non-cancer methylation beta differences, (3) segmentation based on CpG density in regions, (4) dynamic generation of sequencing subregions based on mutual information scores and cancer classification propensity. The analytics system applies selects sequencing subregions and applies a cancer classifier to those subregions to identify cancer presence in the sample.

Systems and methods of the disclosure may include a computer-implemented method, the computer-implemented method including: receiving, at a computer system, nucleic acid sequencing data derived from a methylation assay performed on a biological sample associated with at least one subject; computing, using a processor associated with the computer system, a beta value matrix based on the nucleic acid sequencing data, wherein the beta value matrix comprises one or more missing beta values; addressing, using the processor, the one or more missing beta values in the beta value matrix using a missing beta value completion approach; identifying, using the processor, one or more principal components in the completed beta value matrix; and training, using the one or more principal components in combination with a predetermined set of clinical variables, a classifier to predict a survival outcome for a target subject associated with a disease type.

In comparison to conventional sequencing pileup algorithms, the process described herein generates sequencing pileups that contains additional information not typically reported by conventional algorithms while also consuming fewer computational resources (e.g., time, processing power, and memory). First, each of a FASTA reference genome and BAM sequence read files are converted to an internal representation. This enables the rapid iteration across nucleotide bases of the sequence reads to determine support characteristics that summarize information of nucleic acid molecules corresponding to positions across the reference genome. Next, the support characteristics of positions across the reference genome are stored through a memory allocation process that utilizes a first and a second temporary storage. This enables the convenient freeing of one temporary storage while the other temporary storage is being used.

Methods and systems are disclosed for covariate prediction from methylation features. A system trains methylation state models that are configured to regress one or more methylation features at a genomic region based on covariates for a given sample. The system utilizes the methylation state models to determine information gain of genomic regions in predicting covariates of interest. The system may, based on the information gain, identify covariate-informative genomic regions. The system trains a covariate prediction model using non-cancer training samples with reported covariate label(s) and methylation features at a plurality of covariate-informative genomic regions. The system may deploy the covariate prediction model for sample swap detection. Additionally, the system may utilize prediction(s) from covariate prediction model(s) to serve as a feature to cancer classification.

An analytics system creates a data structure counting strings of methylation vectors from a healthy control group. The analytics system enumerates possibilities of methylation state vectors given a sample fragment from a subject, and calculates probabilities for all possibilities with a Markov chain probability. The analytics system generates a p-value score for the subject's test methylation state vector by summing the calculated probabilities that are less than or equal to the calculated probability of the possibility matching the test methylation state vector. The analytics system determines the test methylation state vector to be anomalously methylated compared to the healthy control group if the p-value score is below a threshold score. With a number of such sample fragments, the analytics system can filter the sample fragments based on each p-value score. The analytics system can run a classification model on the filtered set to predict whether the subject has cancer.

Methods and systems for detecting cancer and/or determining a cancer tissue of origin are disclosed. In some embodiments, a multiclass cancer classifier is disclosed that is trained with a plurality of biological samples containing cfDNA fragments. The analytics system derives a feature vector for each sample, and the multiclass classifier predicts a probability likelihood for each of a plurality of cancer signal origin (CSO) classes. In some embodiments, the plurality of CSO classes include hematological subtypes, including both hematological malignancies and precursor conditions. In one embodiment, non-cancer samples having high prediction score are pruned from the training sample set. In another embodiment, the analytics system stratifies samples according to prediction score and applies binary threshold cutoffs determined for each stratum.

Aspects of the invention relate to methods for tracking patient health by longitudinally tracking genetic variants in patients, such that it is possible to provide a tumor, or mutation, classification signature. Longitudinal tracking improves the ability to detect minimal residual disease (MRD; the small number of cells that remain in the patient after treatment and/or during remission) and/or treatment response at an early stage, both of which can help guide treatment decisions and guard against missing different intra-/inter-tumor responses in a patient.