SOAPec_v2.01 optional trimming

[seyhan]

Hi Everyone, 

What does optional trimming actually do in SOAPec_v2.01? In the supplementary of Luo et al (2012), it is described as removal of fixed reads that remain as errors. The flags are described as following on the website:

-e [int] Set whether trim(1) suspicious region at the end of reads according to Q_value≤2 or not(0), default=0.

-w [int] Set whether trim(1) error bases with Q_value≤2 instead of correct(0) it , default=0.

Here is an stat output from a run I completed on one of my datasets.

Label_1 num_raw_reads: 201317734

Label_2 num_raw_bases: 29628499489

Label_3 num_result_reads: 200669247

Label_4 num_result_bases: 29568329721

Label_5 num_trimmed_reads: 4157500

Label_6 num_trimmed_bases: 33581072

Label_7 num_deleted_reads: 648487

Label_8 num_corrected_bases_by_Fast_method: 10544170

Label_9 num_corrected_bases_by_BBtree_method: 11341354

Label_10 num_corrected_bases_by_two_methods: 21885524

Label_11 low_quality_bases_filter_ratio: 0.00203081

Label_12 estimated_raw_base_error_ratio: 0.000740168

Does the trimmed reads include the deleted reads or were they discarded in addition to deleted reads? I would appreciate if someone can clarify this?

Many thanks,

Seyhan