Polishing: Glossary

Key Points

Polishing an assembly
  • Short reads have a higher base accuracy than long reads and can be used to remove errors in assemblies generated with long reads.

  • Long reads have a lower accuracy but help generate a more contiguous (less fragmented) assembly, so are used to get the structure of the metagenome, but may have small misassemblies or single nucleotide polymorphisms (SNPs)

  • Medaka is used to polish an assembly with long reads.

  • Pilon is used to polish an assembly with short reads.

QC polished assembly
  • The N50 is the contig length of the 50th percentile, meaning that 50% of the contigs are at least this length in the assembly

  • A misassembly is when a portion of the assembly is incorrectly put back together

  • The largest contig is the longest contiguous piece in the assembly

  • Seqkit can generate summary statistics that will tell us the N50, largest contig and the number of gaps

  • metaQUAST can generate additional information in a report which can be used to identify misassemblies

Glossary

accession
a unique identifier assigned to each sequence or set of sequences
categorical variable
Variables can be classified as categorical (aka, qualitative) or quantitative (aka, numerical). Categorical variables take on a fixed number of values that are names or labels.
cleaned data
data that has been manipulated post-collection to remove errors or inaccuracies, introduce desired formatting changes, or otherwise prepare the data for analysis
conditional formatting
formatting that is applied to a specific cell or range of cells depending on a set of criteria
CSV (comma separated values) format
a plain text file format in which values are separated by commas
factor
a variable that takes on a limited number of possible values (i.e. categorical data)
Gb
gigabyte of file storage or file size
Gbase
a gigabase represents one billion nucleic acid bases (Gbp may indicate one billion base pairs of nucleic acid)
headers
names at tops of columns that are descriptive about the column contents (sometimes optional)
metadata
data which describes other data
NGS
common acronym for “Next Generation Sequencing” currently being replaced by “High Throughput Sequencing”
null value
a value used to record observations missing from a dataset
observation
a single measurement or record of the object being recorded (e.g. the weight of a particular mouse)
plain text
unformatted text
quality assurance
any process which checks data for validity during entry
quality control
any process which removes problematic data from a dataset
raw data
data that has not been manipulated and represents actual recorded values
rich text
formatted text (e.g. text that appears bolded, colored or italicized)
string
a collection of characters (e.g. “thisisastring”)
TSV (tab separated values) format
a plain text file format in which values are separated by tabs
variable
a category of data being collected on the object being recorded (e.g. a mouse’s weight)