mdelrio1
commented
7 years ago @sr320 Hi I have uploaded a version 9 of the big table. I have calculated the expression levels and check if they are female female>male male male>female considering a) female, male=0, female >100 (counts) b) female>male, expression male/ expression female >100 c) male, female=0, male > 100 (counts) d) male>female, expression male/ expression female >0.01
is this clear? I hope so.
@sr320 Hi Steven I think we may have an issue with the expression levels. if you look at the file paper-pano-go/data-results/Male-v-Female-Expression.csv in line 88, you will see a "#NAME?" datum in column "Experiment - Fold Change (original values)"
comp144391_c0_seq2,6024,-6024,#NAME?,462,6024,0,0
or in line comp144300_c0seq10,2,2,âˆ,0,0,0,2
you'll see a "âˆ_". I replaced both with zero in the whole table!!! But, in the first case males have expression (females=0), while in the second case is there reverse (females have expression, males=0). When I obtained the data for the big table I used this column, "Experiment - Fold Change (original values)", to calculate whether the expression value was male, female See notebook paper-pano-go/jupyter-nbs/10Panopea_databases.ipynb In [31]: line 7 "if row["fold_change"]>100" and so on.
This means that we did not consider the contigs where there was a value of zero in the "Female-Total" or "Male-Total" I'm going to run again this withe the original data, but considering: female: ['female_unique']>0, ['male_unique']=0 male: ['female_unique']=0, ['male_unique']>0 female>male ['female_unique']/ ['male_unique']>100 male>female ['female_unique']/ ['male_unique']<.01