@Peter Youa re correct, seems to be a IGV error

@terdon I am have tried viewing the gbk file in IGV which is what I have shown above. the gbk and gff contain all the original CDS features but I have just added in the UTRs. here is the complete gbk file I am using: pastebin.com/8jm2LEk2

@terdon. I have tried with both gbk and gff3. What is pasted is just an excerpt from the gbk file (I will add gff also). The format is just that produced by biopython so I am unsure about it being "no known format" as I don't know what the format should look like. Yes that is what i am using :)

@terdon I have looked at the gff3 file, I am unsure what to expect as I don't know how utr regions are represented but my wish is to see the urt regions represented as a features the same as a CDS. Instead I just see nothing but the normal CDS features

@StupidWolf Is there some way I can share the data with you? If the substrate did not exist in that sample it was simply given a count of 0, so i did fill for missing counts (I think?) if I am understanding correctly.

@StupidWolf I am artificially forcing this though but just putting a count of each substrate for each sample. and the samples are from either origin A or B

@StupidWolf I just checked and there are 9 substrates found in origin B that are not in origin B

@StupidWolf I think I get what you mean but that is as are as I am aware, not the case in this data. There might however examples of substrates only found in origin A or B, Could this cause the same effect?

@StupidWolf I see you say it indicates some of the terms are linear correlated, how come the glm function is not throwing up an error then?

Hi @StupidWolf They are some results from a group project yes, I guess someone else in the group is asking the same question, I will find out who and communicate with them to keep you from answering both of us!