change fuzzy position operators for RHS clipped features

[tomkinsc]

this reverses changes from:
1271307#diff-1fff82e70c2795767d937ffd93ccc616

[dpark01]

So... I understand the thinking, but I'll note that for me, tbl2asn errors with the following:

[tbl2asn 25.6] Bad feature table at line 9 of file LASV_NGA_2018_1392-2.tbl
[tbl2asn 25.6] Bad feature table at line 8 of file LASV_NGA_2018_0026-1.tbl
[tbl2asn 25.6] Bad feature table at line 9 of file LASV_NGA_2018_0998-2.tbl
[tbl2asn 25.6] Bad feature table at line 9 of file LASV_NGA_2018_0998-1.tbl

I ran it as tbl2asn -t authors-nga_lasv.sbt -p . -X C -j "organism=Lassa mammarenavirus mol_type=cRNA" and I can provide you the sbt and cmt files offline for testing.

[dpark01]

testfiles.zip

To reproduce: unpack this zip file into an empty directory, copy in all of the tbl files from viral-ngs/test/input/TestFeatureTransfer/lasv/expected (from the ct-feature-clipping-fix branch), and run tbl2asn -t authors-nga_lasv.sbt -p . -X C -j "organism=Lassa mammarenavirus mol_type=cRNA"

[dpark01]

Okay according to the fine print in the documentation:

Locations of partial (incomplete) features are indicated with a ">" or "<" next to the number. In this example, the first gene, CDS, and mRNA all begin upstream of the start of the nucleotide sequence. The "<" symbol indicates that they are 5' partial features. Furthermore, for the protein to translate correctly, the correct reading frame must be indicated with the qualifier "codon_start" on the first CDS. There is no need to indicate the codon_start on complete CDSs, as it is assumed that the translation starts at the first nucleotide of the interval if no codon_start is provided.

So this implies that < does not mean "to the left of" (ie coordinates that are lower integers than this one). It means "5' partial feature" (even if it's the right end of the feature). So the original encoding was correct.