Bioinformatics: Big Data Big Data: Big data refers to data that is so large, fast or complex that it's difficult or impossible to process using traditional methods. · Volume(scale of data) · Variety(different forms of data) · Velocity(Analysis of streaming data) · Veracity(uncertainty of data) Omics: · Omics: The collective technologies used to explore the roles, relationships, and actions of the various types of molecules that make up the cells of an organism. We have a big data problem! Bottleneck has shifted from data generation to analysis and interpretation! Examples of Big Data & Omics: 200GB of data per patient -Processing hundreds or thousands of patient samples a year Different omics data -Goal millions of samples per year -Results need to be acted upon -Next-Generation Sequencing is noisy -Biological confounding factors NGS(Next-Generation Sequencing) Analytics · NGS-based bioinformatics analytics are designed to 1. Convert signals into data(Primary Analysis) 2. Data to interpretable information(Secondary Analysis)
3. Information into actionable knowledge(Tertiary Analysis) Primary Analysis · Highly integrated with the sequencing instruments and associated onboard software. · Function: Converts the raw signals generated by the sequencing instruments into nucleotide bases with associated quality scores, and ultimately, short nucleotide sequences or "reads". · Often comes with the sequencing. Base Quality · Shows the quality of a base. · Each nucleotide has one base quality. · Higher base quality = lower probability of an error. · Read ID: A given ID for a given template we have sequenced. Secondary Analysis · Collection of methods that operate together to detect genomics variations from quality-scored sequence data(FASTQ). FASTQ format is a text-based format for storing both a biological sequence and its corresponding quality scores. Assembly vs. Mapping · De novo assembly involves generating a genome with a minimal a priori genome information. For this method, longer reads are more desirable > improved assembly. · Assembly involves overlapping of reads; the genome is assembled through reads that overlap with each other without a reference genome.
· Increasing read length = improved assembly Mapping · Most applications involve alignment to a reference genome. Each fragment is mapped back to its complementary sequence in the genome. · In mapping(alignment), reads are mapped onto the reference genome. After that, we would try to find the difference between the reference genome and the complimentary sequence. . The reads overlap because they come from different cells . You can assemble from any part of the overlapping reads as long as they can align with each other. Genomic Variations · Single nucleotide variants/polymorphisms(SNV/SNP) - point mutations(Change in single nucleotide of DNA) · Small insertions/deletions(InDel) · Copy number variations(CNV). · Structural variations(SV-more complex changes compared to CNAs) Secondary Analysis(more details): . Covers the collection of methods that take FASTQ file and provide genomic variations · Input of secondary analysis: FASTQ file · Output of secondary analysis: Genomic variations · Variant Calling: Identifying variants in sequencing. SAM & BAM · SAM is a text file