This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Drug target in orthologous genes, how to find correlation

Hi, I have data for a drug on three bacteria species at genomic level. In the example below, gene aa is sensitive to the drug in speceis1 while not in Species2 & 3 (1=sensitive, 0=resistant). I wanted to know if there is any correlation of drug response between these species?

Gene      Species1    Species2    Species3
aa    1               0            0
ab    1               1           0
ac    1               0            0
ad    1               0            1
ae    0               1           0
af    0               0            1

Thanks a lot.

gene genome r

Hi Michael, Thanks a lot for your response. I am sorry for providing incomplete information. The example given is just a representative of my data. The experiment was performed on individual gene knockout strains and there are 123 different genes (and orthologs in other species) which showed either resistance or sensitivity across three species. I converted the result into binary data (1 for resistance and 0 for sensitive) to find a correlation. Just looking at my data, I could also see there is no correlation but I am looking for a statistical interpretation. I provided a bigger example of my data below. Thanks again.

Gene    Species1    Species2    Species3
Gene1   1   0   0
Gene2   1   0   0
Gene3   1   NA  NA
Gene4   1   0   0
Gene5   1   0   0
Gene6   1   0   0
Gene7   1   0   0
Gene8   1   0   0
Gene9   1   0   0
Gene10  1   0   0
Gene11  0   1   0
Gene12  0   1   0
Gene13  0   1   0
Gene14  0   1   NA
Gene15  0   1   0
Gene16  0   0   1
Gene17  0   0   1
Gene18  0   0   1
Gene19  0   0   1

Please use ADD REPLY/ADD COMMENT when responding to existing posts to keep threads logically organized.

Right, should have refreshed before posting. I'm afraid that sentence is the most used line on Biostars.

Please use ADD COMMENT to reply to earlier answers, as such this thread remains logically structured and easy to follow.

Sorry, I modified as suggested but unable to delete my 2nd post.

1 answer

Based on the little and most likely incorrect* information you have given, it is impossible to determine if it is possible to infer a correlation between genotype (that is possibly what you call 'gene aa') and phenotype (resistance: 1 0 0?) somehow.

However, based on only 3 examples you can think of this as a clear 'No'. It is not possible to establish a significant correlation or even hypothesis across species on only 3 examples across 6 genes. Also there might be thousands of other genes that are not taken into account. What if it is gene 'ah' that makes the bacterium resistant?

  • what I mean by little and incorrect: for example 'gene aa is sensitive... in speceis(sic!)'. First, you should not give 'blinded' information, second, the gene is not sensitive, the organism is. Possibly you mean that the gene is a drug target in one species and not in another?

Hi Michael, Thanks a lot for your response. I am sorry for providing incomplete information. The example given is just a representative of my data. The experiment was performed on individual gene knockout strains and there are 123 different genes (and orthologs in other species) which showed either resistance or sensitivity across three species. I converted the result into binary data (1 for resistance and 0 for sensitive) to find a correlation. Just looking at my data, I could also see there is no correlation but I am looking for a statistical interpretation. I provided a bigger example of my data below. Thanks again.

Gene    Species1    Species2    Species3
Gene1   1   0   0
Gene2   1   0   0
Gene3   1   NA  NA
Gene4   1   0   0
Gene5   1   0   0
Gene6   1   0   0
Gene7   1   0   0
Gene8   1   0   0
Gene9   1   0   0
Gene10  1   0   0
Gene11  0   1   0
Gene12  0   1   0
Gene13  0   1   0
Gene14  0   1   NA
Gene15  0   1   0
Gene16  0   0   1
Gene17  0   0   1
Gene18  0   0   1
Gene19  0   0   1

Log in to answer this question.