L4: Genome assembly and annotation

How to get to complete chromosomes

long reads shotgun sequencing

assembly

scaffolding

scaffold

a larger DNA sequence formed by linking together multiple contigs

scaffolding requires -

scaffolding does not add - to your assembly, it just - the pieces you have

longer range information

missing sequences, orders

main scaffolding approaches

• reference-based

• genetic maps

• optical maps

• chromatin conformation capture

chromatin organization in nucleus

fractal globule: DNA is highly organized in chromosomal regions and topologically associated domains (TADs)

Hi-C: chromatin conformation capture

interacting regions are joined and sequenced (with short reads)

more interactions = physical proximity

Long-range Hi-C interaction can - contigs

connect/order

pixel intensity in the matrix indicates how often a pair of genomic positions interact

scaffolding builds - into -

contigs, chromosomes

how to QC contiguity in scaffolds?

scaffold N50: the length of the shortest scaffold for which longer and equal length scaffolds cover at least 50% of the assembly

Hi-C is the high throughput version of -

to study the link between - and -

A compartment?

B compartment?

TADs

chromatin conformation capture

chromatin conformation, genome function

A compartment = transcriptionally active

B compartment = transcriptionally inactive

TADs are genes and enhancers

alleles and genetic variants on the same copy of a chromosome will be - together

inherited

phasing allows for -

phasing: haplotype-aware genome assembler can distinguish which reads belong to the - in - individuals

two chromosome complements are assembled - (haplotypes)

separation of homologous chromosomes

same copy of a chromosome, heterozygous

independently

trio binning

short read data from parental individuals are used to separate long reads from different haplotypes

inbred diploid individuals have - copies of each chromosome, so theres no need for - when assembling genomes

identical, phasing

three complementary approaches to gene annotation

• ab initio

• homology-based

• experimental

ab initio gene annotation

coding genes have distinctive features

bioinformatic software can scan the genome for possible genes

method of predicting genes in a genome by analyzing its intrinsic properties, without using external evidence like protein or RNA sequences

homology-based gene annotation

takes advantage of sequence conservation across species to determine if a gene is “real”

more powerful between closely related species

experimental support for gene annotation

uses available data to identify possible genes such as

• gene expression data

• protein sequence data

• methylation pattern

tandem or interspersed repeats

sequences of various length occurring in multiple copies throughout the genome

transposons

mobile genetic elements, can move or copy themselves across the genomea

automated gene annotation

bioinformatic pipelines combine different lines of evidence

can require extensive manual curation