GeneMark-ES error

[parulvk]

I am trying to run GeneMark-ES on contigs file with repeats masked out.
The file is a hard mask file(i.e. repeats are replaced by 'N'). When I run
genemark: nohup ./contigs.fa.masked > &logfile
I get the error message:
error reading parameter TERM_TGA_MAT
error in model file org_S1.0mtx
Error on system: prediction step
Please let me know how this error could be resolved. Thank you for your help.

[benikaj]

Error on system: prediction step

Hi parulvk, did you ever figure this error out? I am having the same issue and can't figure out to fix it. Any ideas?

[ebioman]

Hi
Having the same issue - but I guess you never figured it out ?

[asperjelly]

Hi All,

I just started getting a similar error. GeneMark-ES was previously working fine on my system, and now am getting some strange behavior resulting in the same "error on prediction step". Here is the ouput of the error:

GC Range: (0,99)
2956 sequences of length 5 used from 2957
total sequences in bp.acc.seq2
Generating model...
TT     0.16 0.45 0.00 0.00 0.00
TC     0.30 0.49 0.00 0.00 0.00
TA     0.37 0.05 1.00 0.00 0.00
TG     0.17 0.01 0.00 0.00 0.00
CT     0.16 0.33 0.00 0.00 0.00
CC     0.30 0.59 0.00 0.00 0.00
CA     0.37 0.08 1.00 0.00 0.00
CG     0.17 0.00 0.00 0.00 0.00
AT     0.16 0.43 0.00 0.00 0.00
AC     0.30 0.48 0.00 0.00 0.00
AA     0.37 0.08 1.00 0.00 0.00
AG     0.17 0.00 0.00 1.00 0.00
GT     0.16 0.32 0.00 0.00 0.22
GC     0.30 0.59 0.00 0.00 0.28
GA     0.37 0.09 1.00 0.00 0.21
GG     0.17 0.00 0.00 0.00 0.28
Done
sorted.intronLD
Can't exec "head": No such file or directory at /home/ngs/programs/gmes/gmes/splice_phase.bp.v2.0 line 64.
Use of uninitialized value $get in pattern match (m//) at /home/ngs/programs/gmes/gmes/splice_phase.bp.v2.0 line 66.
Could not get the max intron length from sorted.intronLD

ERROR :: set_pseudo_counts: invalid count. Possibly you did not use -n with nucleotide data.
EXITING
Can't exec "head": No such file or directory at /home/ngs/programs/gmes/gmes/gibbs.out2len.pl line 44.
Use of uninitialized value $acc in pattern match (m//) at /home/ngs/programs/gmes/gmes/gibbs.out2len.pl line 45.
0 gibbs.r.out.branch.seq
zero order for BP
GC Range: (0,99)
Use of uninitialized value $length in concatenation (.) or string at /home/ngs/programs/gmes/gmes/make_nt_freq.mat line 104.
0 sequences of length  used from 0
total sequences in gibbs.r.out.branch.seq
Generating model...
Use of uninitialized value $length in concatenation (.) or string at /home/ngs/programs/gmes/gmes/make_nt_freq.mat line 110.
Use of uninitialized value $length in subtraction (-) at /home/ngs/programs/gmes/gmes/make_nt_freq.mat line 117.
T    
Use of uninitialized value $length in subtraction (-) at /home/ngs/programs/gmes/gmes/make_nt_freq.mat line 117.
C    
Use of uninitialized value $length in subtraction (-) at /home/ngs/programs/gmes/gmes/make_nt_freq.mat line 117.
A    
Use of uninitialized value $length in subtraction (-) at /home/ngs/programs/gmes/gmes/make_nt_freq.mat line 117.
G    
Done
cat: No match.
Can't exec "head": No such file or directory at /home/ngs/programs/gmes/gmes/scan_bp_get_gt_bp.pl line 48.
Use of uninitialized value $acc in pattern match (m//) at /home/ngs/programs/gmes/gmes/scan_bp_get_gt_bp.pl line 49.
error, no data in 6 - 3000 range
error in file format, org_S4_bp.0mtx
Error on system: prediction step

[SES]

I am trying to run GeneMark-ES on contigs file with repeats masked out.
The file is a hard mask file(i.e. repeats are replaced by 'N'). When I run
genemark: nohup ./contigs.fa.masked > &logfile
I get the error message:
error reading parameter TERM_TGA_MAT
error in model file org_S1.0mtx
Error on system: prediction step
Please let me know how this error could be resolved. Thank you for your help.

I get the exact same error. Unfortunately, this is not a descriptive message and it is kind of impossible to know what this means, especially since there is no documentation. Also, it appears the code was intentionally written to be as cryptic as possible, which makes debugging also not really possible. Maybe the best approach would be to not sink too much time into this program.