Why the resuming tophat2 command gives a strange error of samtools version?

Question

[Manvendra Singh]

Hi All,

I was mapping paired end fastq files by tophat2 on 10 threads, it got stuck during Searching for junctions via segment mapping. Everything seems to be alright; Here is the output

[2015-08-19 00:16:41] Beginning TopHat run (v2.0.8)
-----------------------------------------------
[2015-08-19 00:16:41] Checking for Bowtie
          Bowtie version:     2.0.5.0
[2015-08-19 00:16:41] Checking for Samtools
        Samtools version:     0.1.17.0
[2015-08-19 00:16:42] Checking for Bowtie index files
[2015-08-19 00:16:42] Checking for reference FASTA file
[2015-08-19 00:16:42] Generating SAM header for /home/msingh/RNAseq/Monkey/Genome/Marmoset/index/calJac3
    format:         fastq
    quality scale:     phred33 (default)
[2015-08-19 00:17:37] Reading known junctions from GTF file
[2015-08-19 00:17:43] Preparing reads
     left reads: min. length=101, max. length=101, 98351402 kept reads (18044 discarded)
    right reads: min. length=101, max. length=101, 98364234 kept reads (5212 discarded)
[2015-08-19 01:36:00] Creating transcriptome data files..
[2015-08-19 01:37:12] Building Bowtie index from callithrix_ensembl_genes.fa
[2015-08-19 01:49:24] Mapping left_kept_reads to transcriptome callithrix_ensembl_genes with Bowtie2
[2015-08-19 03:14:04] Mapping right_kept_reads to transcriptome callithrix_ensembl_genes with Bowtie2
[2015-08-19 04:40:49] Resuming TopHat pipeline with unmapped reads
[2015-08-19 04:40:50] Mapping left_kept_reads.m2g_um to genome calJac3 with Bowtie2
[2015-08-20 09:40:53] Mapping left_kept_reads.m2g_um_seg1 to genome calJac3 with Bowtie2 (1/4)
[2015-08-20 11:04:05] Mapping left_kept_reads.m2g_um_seg2 to genome calJac3 with Bowtie2 (2/4)
[2015-08-20 12:41:12] Mapping left_kept_reads.m2g_um_seg3 to genome calJac3 with Bowtie2 (3/4)
[2015-08-20 14:13:16] Mapping left_kept_reads.m2g_um_seg4 to genome calJac3 with Bowtie2 (4/4)
[2015-08-20 16:30:35] Mapping right_kept_reads.m2g_um to genome calJac3 with Bowtie2
[2015-08-22 01:27:44] Mapping right_kept_reads.m2g_um_seg1 to genome calJac3 with Bowtie2 (1/4)
[2015-08-22 02:55:12] Mapping right_kept_reads.m2g_um_seg2 to genome calJac3 with Bowtie2 (2/4)
[2015-08-22 04:30:08] Mapping right_kept_reads.m2g_um_seg3 to genome calJac3 with Bowtie2 (3/4)
[2015-08-22 06:01:47] Mapping right_kept_reads.m2g_um_seg4 to genome calJac3 with Bowtie2 (4/4)
[2015-08-22 08:20:15] Searching for junctions via segment mapping

Then I tried to resume it

by

/usr/local/bin/tophat2 -R /path_to_output

Now it is showing an error as

[2015-08-24 14:46:42] Resuming TopHat run in directory '/home/msingh/RNAseq/Monkey/mapresults/Callithrix/iPSC/' stage 'find_juncs'
-----------------------------------------------
[2015-08-24 14:46:42] Checking for Bowtie
          Bowtie version:     2.0.5.0
[2015-08-24 14:46:42] Checking for Samtools
Traceback (most recent call last):
  File "/usr/local/tophat-2.0.8/tophat", line 4030, in <module>
    sys.exit(main())
  File "/usr/local/tophat-2.0.8/tophat", line 3831, in main
    check_samtools()
  File "/usr/local/tophat-2.0.8/tophat", line 1545, in check_samtools
    samtools_version_str, samtools_version_arr = get_samtools_version()
  File "/usr/local/tophat-2.0.8/tophat", line 1527, in get_samtools_version
    samtools_version_arr = [int(version_match.group(x)) for x in [1,2,3]]
AttributeError: 'NoneType' object has no attribute 'group'`

Now, as its shown here, -R is showing samtools version error, whereas, when I gave 1st shot to tophat2, then there was no issue with samtools

Does tophat2 requires another version of samtools for finding junctions, when we Resume it?

Please help, After 3 days of running command it can not resume makes me little frustrated, should I switch to STAR for everything?