linux shell script which can do this task

Question

[Confused_human]

HI everyone,

I have some dataset like this :

column1 - protein IDs

column2 - domain

Agarivorans_aestuarii_peg_2571  NODOMAIN
Agarivorans_aestuarii_peg_2572  Wzy_C   
Agarivorans_aestuarii_peg_2573  NODOMAIN
Agarivorans_aestuarii_peg_2574  Polysacc_synt_C 
Agarivorans_aestuarii_peg_2575  NODOMAIN
Agarivorans_aestuarii_peg_2576  Caps_syn_GfcC_N Caps_syn_GfcC_C
Agarivorans_aestuarii_peg_2577  Polysacc_synt_2 
Agarivorans_aestuarii_peg_2578  NODOMAIN
--
Aliivibrio_salmonicida_peg_0270 Wzy_C   
Aliivibrio_salmonicida_peg_0271         NODOMAIN
Aliivibrio_salmonicida_peg_0272       NODOMAIN
Aliivibrio_salmonicida_peg_0273        NODOMAIN
Aliivibrio_salmonicida_peg_0274      Caps_syn_GfcC_N Caps_syn_GfcC_C
Aliivibrio_salmonicida_peg_0275        NODOMAIN
Aliivibrio_salmonicida_peg_0278       Poly_export     Caps_syn_GfcC_C
Aliivibrio_salmonicida_peg_079         Wzz     
Aliivibrio_salmonicida_peg_0280        NODOMAIN

Output should look like :

Agarivorans_aestuarii_peg_2572  Wzy_C   
Agarivorans_aestuarii_peg_2573  NODOMAIN
Agarivorans_aestuarii_peg_2574  Polysacc_synt_C 
Agarivorans_aestuarii_peg_2575  NODOMAIN
Agarivorans_aestuarii_peg_2576  Caps_syn_GfcC_N Caps_syn_GfcC_C
Agarivorans_aestuarii_peg_2577  Polysacc_synt_2 
--
Aliivibrio_salmonicida_peg_0270 Wzy_C   
Aliivibrio_salmonicida_peg_0271         NODOMAIN
Aliivibrio_salmonicida_peg_0272       NODOMAIN
Aliivibrio_salmonicida_peg_0273        NODOMAIN
Aliivibrio_salmonicida_peg_0274      Caps_syn_GfcC_N Caps_syn_GfcC_C
Aliivibrio_salmonicida_peg_0275        NODOMAIN
Aliivibrio_salmonicida_peg_0278       Poly_export     Caps_syn_GfcC_C
Aliivibrio_salmonicida_peg_079         Wzz

Everything that is coming in between two known domains will get printed and everything else will get removed.

and if in it is like this :

Atlantibacter_hermannii_peg_1261        NODOMAIN
Atlantibacter_hermannii_peg_1262        NODOMAIN
Atlantibacter_hermannii_peg_1263        NODOMAIN
Atlantibacter_hermannii_peg_1264        Wzz
Atlantibacter_hermannii_peg_1265        NODOMAIN

then it will give output :

Atlantibacter_hermannii_peg_1264        Wzz

Please tell me how this can be done with linux shell scripting or any other script.

I have tried sort and uniq but it did not work fo this as the protein peg IDs are different.

and there is a partitioning line like "--" between each cluster like the way I have showed in the above example data.

Thank you

[cpad0112]

Post is confusing. Can you rephrase it with smaller example and with only one query per post?

[Confused_human]

OKAY

So my dataset has two columns

1st col - protein IDs

2nd col - domain

input -

Abiotrophia_defectiva_peg_1828  NODOMAIN
Abiotrophia_defectiva_peg_1829  wzz
Abiotrophia_defectiva_peg_1830  NODOMAIN
Abiotrophia_defectiva_peg_1831  wzx
Abiotrophia_defectiva_peg_1832 wza
Abiotrophia_defectiva_peg_1833 NODOMAIN

Output -

Abiotrophia_defectiva_peg_1829  wzz
Abiotrophia_defectiva_peg_1830  NODOMAIN
Abiotrophia_defectiva_peg_1831  wzx
Abiotrophia_defectiva_peg_1832 wza

it will print all the entries in between two known domains. like for this case everything in between wzz and wza will get printed .

this should be applied for all the entries .

and these protein clusters are separated by "--"

and if its like this -

Atlantibacter_hermannii_peg_1262 NODOMAIN
Atlantibacter_hermannii_peg_1263 NODOMAIN
Atlantibacter_hermannii_peg_1264 Wzz
Atlantibacter_hermannii_peg_1265 NODOMAIN

it will give output

Atlantibacter_hermannii_peg_1264 Wzz

because only one known domain is there in that cluster