error in handling of identity exceptions

- enzymes with high sequence identity could be specified to use higher
  blast identity cutoff (dat/exception.tbl)
- wrong handling could happen because the comment column was also used
  for ec/reaction matching
- fix only consideres first column (ec number or reaction name)
- this affects the alcohol dehydrogenase (1.1.1.1) which is mentioned in
  the comment column of the exception file and was not found by blast search
  for example for Enterococcus faecalis because the higher identity cutoff
  (70%) was needed

Author: jotech

dat/exception.tbl

@@ -1,16 +1,16 @@
 # file with enzymes having a high similar sequence identity to differently working enzymes (false friends) => need more strict blast search
-enzyme/reaction comment
-1.9.3.1 cytochrome-c oxidase
-4.1.1.39 RuBisCo
-D-ribulose-1,5-bisphosphate cleaving dioxygenase    RuBisCo
-1.9.6.1 nitrate reductase nap
-1.7.2.5 nitric oxide reductase
-4.1.2.22 fructose-6-phosphate phosphoketolase
-2.6.1.96 gamma aminobutyrate transaminase
-2.8.3.5 succinyl-CoA:acetoacetate CoA-transferase
-2.3.1.31 homoserine O-acetyltransferaseInferred from experiment
-3.2.1.51 alpha-L-fucosidase
-4.1.1.17 ornithine decarboxylase (high similarity with other protein: 10.1074/jbc.R500031200)
-1.1.1.244 methanol dehydrogenase (high similarity to EC 1.1.1.1)
-1.1.1.26 glycolate dehydrogenase
-2.6.1.19 4-aminobutyrate aminotransferase
+enzyme/reaction	comment
+1.9.3.1	cytochrome-c oxidase
+4.1.1.39	RuBisCo
+D-ribulose-1,5-bisphosphate cleaving dioxygenase	RuBisCo
+1.9.6.1	nitrate reductase nap
+1.7.2.5	nitric oxide reductase
+4.1.2.22	fructose-6-phosphate phosphoketolase
+2.6.1.96	gamma aminobutyrate transaminase
+2.8.3.5	succinyl-CoA:acetoacetate CoA-transferase
+2.3.1.31	homoserine O-acetyltransferaseInferred from experiment
+3.2.1.51	alpha-L-fucosidase
+4.1.1.17	ornithine decarboxylase (high similarity with other protein: 10.1074/jbc.R500031200)
+1.1.1.244	methanol dehydrogenase (high similarity to EC 1.1.1.1)
+1.1.1.26	glycolate dehydrogenase
+2.6.1.19	4-aminobutyrate aminotransferase

src/gapseq_find.sh

@@ -592,9 +592,9 @@ do
         geneRef=$(grep -wFe $rea $metaGenes | awk -vFPAT='([^,]*)|("[^"]+")' -vOFS=, {'print $5'})
         [[ verbose -ge 1 ]] && echo -e "\t$j) $rea $reaName $ec" $geneName
         [[ -z "$rea" ]] && { continue; }
-        [[ -n "$ec" ]] && [[ -n "$reaName" ]] && [[ -n "$EC_test" ]] && { is_exception=$(grep -Fw -e "$ec" -e "$reaName" $dir/../dat/exception.tbl | wc -l); }
-        ( [[ -z "$ec" ]] || [[ -z "$EC_test" ]] ) && [[ -n "$reaName" ]] && { is_exception=$(grep -Fw "$reaName" $dir/../dat/exception.tbl | wc -l); }
-        [[ -n "$ec" ]] && [[ -z "$reaName" ]] && [[ -n "$EC_test" ]] && { is_exception=$(grep -Fw "$ec" $dir/../dat/exception.tbl | wc -l); }
+        [[ -n "$ec" ]] && [[ -n "$reaName" ]] && [[ -n "$EC_test" ]] && { is_exception=$(cat $dir/../dat/exception.tbl | cut -f 1 | grep -Fw -e "$ec" -e "$reaName" | wc -l); }
+        ( [[ -z "$ec" ]] || [[ -z "$EC_test" ]] ) && [[ -n "$reaName" ]] && { is_exception=$(cat $dir/../dat/exception.tbl | cut -f 1 | grep -Fw "$reaName" | wc -l); }
+        [[ -n "$ec" ]] && [[ -z "$reaName" ]] && [[ -n "$EC_test" ]] && { is_exception=$(cat $dir/../dat/exception.tbl | cut -f 1 | grep -Fw "$ec" | wc -l); }
         if [[ $is_exception -gt 0 ]] && [[ $identcutoff -lt $identcutoff_exception ]];then # take care of similair enzymes with different function
             identcutoff_tmp=$identcutoff_exception
             [[ verbose -ge 1 ]] && echo -e "\t\tUsing higher identity cutoff for $rea"