How to prepare a table of last exons

Question

[lauren.fehrman]

I downloaded a bed file of coding exons from NCBI Table Browser. chr1 67000041 67000051 NM_001308203_cds_1_0_chr1_67000042_f 0 + chr1 67091529 67091593 NM_001308203_cds_2_0_chr1_67091530_f 0 + chr1 67098752 67098777 NM_001308203_cds_3_0_chr1_67098753_f 0 + chr1 67105459 67105516 NM_001308203_cds_4_0_chr1_67105460_f 0 + chr1 67108492 67108547 NM_001308203_cds_5_0_chr1_67108493_f 0 + chr1 67109226 67109402 NM_001308203_cds_6_0_chr1_67109227_f 0 + chr1 67136677 67136702 NM_001308203_cds_7_0_chr1_67136678_f 0 + chr1 67137626 67137678 NM_001308203_cds_8_0_chr1_67137627_f 0 + chr1 67138963 67139049 NM_001308203_cds_9_0_chr1_67138964_f 0 + chr1 67142686 67142779 NM_001308203_cds_10_0_chr1_67142687_f 0 + chr1 67145360 67145435 NM_001308203_cds_11_0_chr1_67145361_f 0 + chr1 67154830 67154958 NM_001308203_cds_12_0_chr1_67154831_f 0 + chr1 67155872 67155999 NM_001308203_cds_13_0_chr1_67155873_f 0 + chr1 67160121 67160187 NM_001308203_cds_14_0_chr1_67160122_f 0 + chr1 67184976 67185088 NM_001308203_cds_15_0_chr1_67184977_f 0 + chr1 67194946 67195102 NM_001308203_cds_16_0_chr1_67194947_f 0 + chr1 67199430 67199563 NM_001308203_cds_17_0_chr1_67199431_f 0 + chr1 67205017 67205220 NM_001308203_cds_18_0_chr1_67205018_f 0 + chr1 67206340 67206405 NM_001308203_cds_19_0_chr1_67206341_f 0 + chr1 67206954 67207119 NM_001308203_cds_20_0_chr1_67206955_f 0 + chr1 67208755 67208778 NM_001308203_cds_21_0_chr1_67208756_f 0 + chr1 67000041 67000051 NM_032291_cds_0_0_chr1_67000042_f 0 + chr1 67091529 67091593 NM_032291_cds_1_0_chr1_67091530_f 0 + chr1 67098752 67098777 NM_032291_cds_2_0_chr1_67098753_f 0 + chr1 67101626 67101698 NM_032291_cds_3_0_chr1_67101627_f 0 + chr1 67105459 67105516 NM_032291_cds_4_0_chr1_67105460_f 0 + chr1 67108492 67108547 NM_032291_cds_5_0_chr1_67108493_f 0 + chr1 67109226 67109402 NM_032291_cds_6_0_chr1_67109227_f 0 + chr1 67126195 67126207 NM_032291_cds_7_0_chr1_67126196_f 0 +

What I am interested in is a bed file containing only the last exon for each gene along with the gene identifier information. For example: chr1 67208755 67208778 NM_001308203 0 +

chr1 67126195 67126207 NM_032291 0 +

I'm pretty inexperienced with awk, but I'm hoping there is a straightforward command to do this

Thanks in advance,

Lauren

Answer 1

[Pierre Lindenbaum]

for all '+' exons;

grep '+$' intput.tsv | sed 's/_cds/\t/' | sort -t $'\t' -k4,4 -k1,1 -k2,2rn | cut -f 1-4 | sort -t $'\t' -k4,4 -k1,1 --stable --uniq

for '-' exons:

grep -- '-$' intput.tsv | sed 's/_cds/\t/' | sort -t $'\t' -k4,4 -k1,1 -k2,2n | cut -f 1-4 | sort -t $'\t' -k4,4 -k1,1 --stable --uniq

(very similar to how to remove rows based on certain characters )