How To Extract Max Score Blast Hit Among A Data Set?

Question

[HG]

Hi everyone, I performed a local blastp with a query protein vs 5 genome (ORF data ). Now i want to extract all the top 5 sequence automatically and make a final top hit file through awk/perl/shell. Can any one please help me out how to do it.

    # BLASTP 2.2.25 [Feb-01-2011]
# Query: gnl|ECOLI|EG10158-MONOMER ClpP 455901..456524 Escherichia coli K-12 substr. MG1655
# Database: mybd
# Fields: Query id, Subject id, % identity, alignment length, mismatches, gap openings, q. start, q. end, s. start, s. end, e-value, bit score
gnl|ECOLI|EG10158-MONOMER    V161_10_length_174349_cov_20.0403_ID_19_63    100.00    207    0    0    1    207    1    207    9e-157     431
gnl|ECOLI|EG10158-MONOMER    RS374_6_length_260428_cov_19.5118_ID_11_179    100.00    207    0    0    1    207    1    207    9e-157     431
gnl|ECOLI|EG10158-MONOMER    H127_1_length_711606_cov_31.5057_ID_1_431    100.00    207    0    0    1    207    1    207    9e-157     431
gnl|ECOLI|EG10158-MONOMER    sff_11_length_151555_cov_28.2432_ID_21_27    100.00    207    0    0    1    207    1    207    9e-157     431
gnl|ECOLI|EG10158-MONOMER    H75_1_length_711552_cov_31.2637_ID_1_431    100.00    207    0    0    1    207    1    207    9e-157     431
gnl|ECOLI|EG10158-MONOMER    H75_26_length_34692_cov_38.2655_ID_51_32    28.24    131    90    2    73    202    63    190    3e-11    60.8
gnl|ECOLI|EG10158-MONOMER    RS374_42_length_11164_cov_52.9994_ID_83_6    28.24    131    90    2    73    202    63    190    4e-11    60.8
gnl|ECOLI|EG10158-MONOMER    H127_2_length_512186_cov_33.056_ID_3_271    26.05    119    67    5    22    121    302    418    5.6    26.9
gnl|ECOLI|EG10158-MONOMER    H75_3_length_319828_cov_30.2145_ID_5_224    26.05    119    67    5    22    121    302    418    5.6    26.9
gnl|ECOLI|EG10158-MONOMER    V161_3_length_303745_cov_18.9698_ID_5_105    26.05    119    67    5    22    121    302    418    6.3    26.9
gnl|ECOLI|EG10158-MONOMER    RS374_2_length_370043_cov_19.1337_ID_3_206    26.27    118    66    5    23    121    303    418    6.4    26.9
gnl|ECOLI|EG10158-MONOMER    sff_1_length_375148_cov_28.8962_ID_1_190    26.27    118    66    5    23    121    303    418    6.4    26.9
gnl|ECOLI|EG10158-MONOMER    RS374_4_length_285072_cov_19.5222_ID_7_219    30.30    33    23    0    128    160    176    208    7.2    26.6

[PoGibas]

Have you tried solving this yourself?
How can I sort blast output based on e-value?
I would use linux sort and head.

[HG]

I appreciate sort and head are useful command. but for my purpose i dont know how much it will be useful. In context of my question if you got properly i need top 5 max score sequence in a separate file so i can pass those sequences in next phase of my pipe.
for example i need : star position to end position of each top 5 sequences.

[User000]

Getting top 10 sequences of BLAST results Bio Python have a look here. Also, you may use python script taking only top 5 scores

Answer 1

[5heikki]

sort -k1,1 -k12,12gr -k11,11g output | sort -u -k1,1 --merge > best
grep -v -f best output > outputWithoutBest
sort -k1,1 -k12,12gr -k11,11g outputWithoutBest | sort -u -k1,1 --merge > secondBest
etc.
cat best secondBest .. | sort -k1,1 > best5

There's probably a one-liner but I'm too lazy to construct it. Also, stop being lazy. Google would have known this..