Ray Error: miscount of the number of reads

[standonn]

Dear all,

I am trying to run Ray to de novo assemble a nematode genome.
I run into the following error:

mpirun \
-n 32 \
/mnt/Programs/Ray-2.3.1/Ray \
-k 81 \
-o rayk81 \
-p ../../clean_reads/PE_1.noCont.ec.fa ../../clean_reads/PE_2.noCont.ec.fa \
-p ../../clean_reads/MP3_1.noCont.ec.fa ../../clean_reads/MP3_2.noCont.ec.fa \
-p ../../clean_reads/MP5_1.noCont.ec.fa ../../clean_reads/MP5_2.noCont.ec.fa \
-p ../../clean_reads/MP8_1.noCont.ec.fa ../../clean_reads/MP8_2.noCont.ec.fa

[.....]

Rank 7: File ../../clean_reads/MP8_2.noCont.ec.fa (Number 7) has 10233322 sequences
Rank 6: File ../../clean_reads/MP8_1.noCont.ec.fa (Number 6) has 10231913 sequences
Rank 5: File ../../clean_reads/MP5_2.noCont.ec.fa (Number 5) has 10722610 sequences
Rank 2: File ../../clean_reads/MP3_1.noCont.ec.fa (Number 2) has 14151655 sequences
Rank 4: File ../../clean_reads/MP5_1.noCont.ec.fa (Number 4) has 10722031 sequences
Rank 3: File ../../clean_reads/MP3_2.noCont.ec.fa (Number 3) has 14152522 sequences
Rank 0: File ../../clean_reads/PE_1.noCont.ec.fa (Number 0) has 100860164 sequences
Rank 1: File ../../clean_reads/PE_2.noCont.ec.fa (Number 1) has 100860164 sequences
Rank 0 wrote rayk81/NumberOfSequences.txt
Rank 0 wrote rayk81/SequencePartition.txt

Rank 0 : Error, ../../clean_reads/MP3_1.noCont.ec.fa contains 14151655 sequences and ../../clean_reads/MP3_2.noCont.ec.fa contains 14152522 sequences (must be the same)

The problem detected by Ray (not the same number of sequences in the left and right read files) is wrong. Actually Ray does not seem to correctly count the sequences:

grep -c '^>' ../../clean_reads/MP3_1.noCont.ec.fa
9763950
grep -c '^>' ../../clean_reads/MP3_2.noCont.ec.fa  
9763950

A head of my read files looks completely normal (regular multiline fasta).
I have also run other assemblers successfully on this data, so I know there is no format problem with the files.

Any insights on what could be causing this problem?

Best Wishes,
Sophie