Support SAM normalization

[totakke]

No description provided.

[codecov]

Codecov Report

Merging #86 into master will increase coverage by <.01%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master      #86      +/-   ##
==========================================
+ Coverage   82.63%   82.63%   +<.01%     
==========================================
  Files          60       60              
  Lines        4140     4146       +6     
  Branches      432      434       +2     
==========================================
+ Hits         3421     3426       +5     
+ Misses        287      286       -1     
- Partials      432      434       +2

Impacted Files	Coverage Δ
src/cljam/algo/normal.clj	80.95% <66.66%> (+0.95%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6da224e...6abf4b0. Read the comment docs.

[alumi]

Thank you for the patch!

I think we need some more changes.

The problems are

• sam->bam or bam->sam cannot be converted by read-blocks/write-blocks. cljam.algo.sorter switches read/write functions according to the combination of input/output formats.
• bam normalization can let blocks pass through because read-blocks keeps ref-id without referring real RNAME, but we have to update RNAME for sam.

For example, following input

@SQ	SN:1	LN:45
@SQ	SN:2	LN:40
r003	16	1	29	30	6H5M	*	0	0	TAGGC	*
r001	163	1	7	30	8M4I4M1D3M	=	37	39	TTAGATAAAGAGGATACTG	*	XX:B:S,12561,2,20,112
r002	0	1	9	30	1S2I6M1P1I1P1I4M2I	*	0	0	AAAAGATAAGGGATAAA	*
r003	0	1	9	30	5H6M	*	0	0	AGCTAA	*
x3	0	2	6	30	9M4I13M	*	0	0	ttataaaacAAATaattaagtctaca	??????????????????????????
r004	0	1	16	30	6M14N1I5M	*	0	0	ATAGCTCTCAGC	*
r001	83	1	37	30	9M	=	7	-39	CAGCGCCAT	*
x1	0	2	1	30	20M	*	0	0	aggttttataaaacaaataa	????????????????????
x2	0	2	2	30	21M	*	0	0	ggttttataaaacaaataatt	?????????????????????
x4	0	2	10	30	25M	*	0	0	CaaaTaattaagtctacagagcaac	?????????????????????????
x6	0	2	14	30	23M	*	0	0	Taattaagtctacagagcaacta	???????????????????????
x5	0	2	12	30	24M	*	0	0	aaTaattaagtctacagagcaact	????????????????????????

gives this broken output sam with the current implementation.

@SQ     SN:chr1 LN:45
@SQ     SN:chr2 LN:40
r003    16      1       29      30      6H5M    *       0       0       TAGGC   *
r001    163     1       7       30      8M4I4M1D3M      =       37      39      TTAGATAAAGAGGATACTG     *       XX:B:S,12561,2,20,112
r002    0       1       9       30      1S2I6M1P1I1P1I4M2I      *       0       0       AAAAGATAAGGGATAAA       *
r003    0       1       9       30      5H6M    *       0       0       AGCTAA  *
x3      0       2       6       30      9M4I13M *       0       0       ttataaaacAAATaattaagtctaca      ??????????????????????????
r004    0       1       16      30      6M14N1I5M       *       0       0       ATAGCTCTCAGC    *
r001    83      1       37      30      9M      =       7       -39     CAGCGCCAT       *
x1      0       2       1       30      20M     *       0       0       aggttttataaaacaaataa    ????????????????????
x2      0       2       2       30      21M     *       0       0       ggttttataaaacaaataatt   ?????????????????????
x4      0       2       10      30      25M     *       0       0       CaaaTaattaagtctacagagcaac       ?????????????????????????
x6      0       2       14      30      23M     *       0       0       Taattaagtctacagagcaacta ???????????????????????
x5      0       2       12      30      24M     *       0       0       aaTaattaagtctacagagcaact        ????????????????????????

We get NPE if we specify bam for output.

In conclusion, we'd better use read-alignments/write-alignments if either of the input/output formats is sam.

[totakke]

I overlooked that point. Thank you for advice. I'll fix it.

[alumi]

LGTM! Thank you 👍