5 releases
Uses old Rust 2015
| 0.3.1 | Sep 14, 2017 |
|---|---|
| 0.3.0 | Jul 20, 2017 |
| 0.2.2 | Jul 18, 2017 |
| 0.2.1 | Jul 18, 2017 |
| 0.2.0 | Jul 18, 2017 |
#6 in #gff
34KB
744 lines
Intersections
This program finds the overlap of sequences and genes using format 6 blastn output (http://www.metagenomics.wiki/tools/blast/blastn-output-format-6)
qseqid sseqid pident length mismatch gapopen qstart qend sstart send evalue bitscore
Query_1 accn|JISN01000002 100.000 28 0 0 29 56 37930 37957 1.32e-08 52.8
and gff3 output (from prokka)
##gff-version 3
##sequence-region accn_JISN01000001 1 334949
...
accn_JISN01000001 Prodigal:2.6 CDS 240 2849 . + 0 ID=NKHGEDLF_00001;Name=clpB;gene=clpB;inference=ab initio prediction:Prodigal:2.6,similar to AA sequence:UniProtKB:Q7A6G6;locus_tag=NKHGEDLF_00001;product=Chaperone protein ClpB
...
>accn_JISN01000001
AATTAATTATCGACCAAGAAAGTGTTTAAATTGGAAGTTTCCTTATGAAGTTTTAT
...
Lines 9 and 10 of the blastn output are compared to lines 4 and 5 of the gff3 file (section type 2) for overlap. Any number of bla files can be intersected with an equal number of MATCHING gff files.
Prerequisites
Folder of .bla files and .gff files MATCHED by NAME (I.E. genome1.bla genome1.gff genome2.bla genome2.gff). Bla files are files created in blastn format 6 by the blasting of one or more sequences against the respective genome. Gff3 files are created (for example) by prokka v1.12 (http://www.vicbioinformatics.com/software.prokka.shtml) for a respective genome.
Installing
First download rust (instructions from https://rustup.rs/)
curl https://sh.rustup.rs -sSf | sh
Then download the crate for intersections
cargo +nightly install sequence-intersections
Intersections can then be found in ~/.cargo/bin/ If a previous version of intersections already exists in the directory use
cargo +nightly install -f sequence-intersections
Output and Options
| Column | Description |
|---|---|
| name | Name of gene according to gff file. Regions between two genes are denoted Between(GeneNameBefore, GeneNameAfter). Hypothetical proteins are denoted HypotheticalAfter(GeneName) or HypotheticalBefore(GeneName) |
| product | Product of gene according to gff file. Same style as name. |
| total_overlap | Amount of sequence which intersected at this gene. If a sequence of 31 in the blast in put file completely overlapped with this gene (IE blast was in ID_1 and spanned 1000-1031 and the gene was in ID_1 and spanned 1000-1500) then the total_overlap for this gene would add +31. |
| genome_count | The number of genomes which had at least one sequence overlap this gene with at least 1 total_overlap. |
| start_avg | The average start for this gene according to the gff file. |
| start_stdev | The standard deviation of the start of this gene. |
| end_avg | The average end for this gene according to the gff file. |
| end_stdev | The standard deviation of the end of this gene. |
| length_avg | The average span of each gene (# of nucleotides long). Is not related to start or end location but only length of the gene. |
Example
Example blast and gff intersections at: https://github.com/dUmich/intersections-example
Errors
Run with this command preceding to get warnings
RUST_LOG=warn
Built with
Versioning
Authors
- Lee Bousfield - Free-lance code wizard - PlasmaPower
- Daniel Harris - Researcher, Snitkin Lab, University of Michigan - dUmich
Dependencies
~7MB
~131K SLoC