Skip to main content
Fig. 1 | Genome Biology

Fig. 1

From: Bystro: rapid online variant annotation and natural-language filtering at whole-genome scale

Fig. 1

Using Bystro online to find alleles of interest in sequencing experiments. a After logging in (https://bystro.io/), users upload one or more VCF or SNP-format files—containing alleles from a sequencing experiment—from a computer or a connected Amazon S3 bucket. Datasets of over 890 GB, containing thousands of samples and tens of millions of variants, are supported. The data are rapidly annotated in the cloud, using descriptions from public sources (e.g. RefSeq, dbSNP, Clinvar, and others). The annotated results can be filtered using Bystro’s natural-language search engine and any search results can be saved as new annotations. Annotated experiments and saved results can be viewed online, downloaded as tab-delimited text, or uploaded back to linked Amazon S3 buckets. b An example of using Bystro’s natural-language search engine to filter 1000 Genomes Phase 3 (https://bystro.io/public). To do so, users may type natural phrases, specific terms, numerical ranges, or apply filters on any annotated field. Queries are flexible, allowing misspelled terms such as “earl-onset” to accurately match. Complex tasks, such as identifying de novo variants can be achieved by using Boolean operators (AND, OR, NOT, +, -), exact-match filters, and user-defined terms. For instance, after labeling the “proband” and their “parents,” the user could simply search probandparents or combine with additional parameters for more refined queries, i.e. probandparents missingness < .1 gnomad.exomes.af_nfe < .001

Back to article page