tell me the different between two version cap?

[robertorun]

Recently , i have used two version cap3, the before version is buided at 04/15/05, the latest VersionDate is 10/15/07.

when i used the two version software to assemble the following datasets with default parameters, but, i get different result, the old version products a contigs, however, the latest result is two singletons.

So, i compare the parameters of two versions.I find, there are four different default parameters just bold line, and at the new version, three parameters (just red line)are added.

My question is that, who can tell me the three new added parameters mean ?

The another puzzle is that, i want to give the old parameter to new version, the command is "cap3 ./temp.fas -o 40 -p 80 -s 900 -y 250", but the result is also two singletons. Why?????? Is it the three new para meters to effect the assmmble???

Best regards!

old parameters(04/15/05)
-a N specify band expansion size N > 10 (20)
-b N specify base quality cutoff for differences N > 15 (20)
-c N specify base quality cutoff for clipping N > 5 (12)
-d N specify max qscore sum at differences N > 20 (200)
-e N specify clearance between no. of diff N > 10 (30)
-f N specify max gap length in any overlap N > 1 (20)
-g N specify gap penalty factor N > 0 (6)
-h N specify max overhang percent length N > 2 (20)
-m N specify match score factor N > 0 (2)
-n N specify mismatch score factor N < 0 (-5)
-o N specify overlap length cutoff > 20 (40)
-p N specify overlap percent identity cutoff N > 65 (80)
-r N specify reverse orientation value N >= 0 (1)
-s N specify overlap similarity score cutoff N > 400 (900)
-t N specify max number of word matches N > 30 (300)
-u N specify min number of constraints for correction N > 0 (3)
-v N specify min number of constraints for linking N > 0 (2)
-w N specify file name for clipping information (none)
-x N specify prefix string for output file names (cap)
-y N specify clipping range N > 5 (250)
-z N specify min no. of good reads at clip pos N > 0 (3)

new parameters(10/15/07)
-a N specify band expansion size N > 10 (20)
-b N specify base quality cutoff for differences N > 15 (20)
-c N specify base quality cutoff for clipping N > 5 (12)
-d N specify max qscore sum at differences N > 20 (200)
-e N specify clearance between no. of diff N > 10 (30)
-f N specify max gap length in any overlap N > 1 (20)
-g N specify gap penalty factor N > 0 (6)
-h N specify max overhang percent length N > 2 (20)
-i N specify segment pair score cutoff N > 20 (40)
-j N specify chain score cutoff N > 30 (80)
-k N specify end clipping flag N >= 0 (1)
-m N specify match score factor N > 0 (2)
-n N specify mismatch score factor N < 0 (-5)
-o N specify overlap length cutoff > 15 (40)
-p N specify overlap percent identity cutoff N > 65 (90)
-r N specify reverse orientation value N >= 0 (1)
-s N specify overlap similarity score cutoff N > 250 (900)
-t N specify max number of word matches N > 30 (300)
-u N specify min number of constraints for correction N > 0 (3)
-v N specify min number of constraints for linking N > 0 (2)
-w N specify file name for clipping information (none)
-x N specify prefix string for output file names (cap)
-y N specify clipping range N > 5 (100)
-z N specify min no. of good reads at clip pos N > 0 (3)




>lcl|Seq274836 No definition line found
TCAGCCGCGCAGGTATACTGACAGTGATATCACTTCCTACTAGCTAGCTGCTACTTGAAA
CTAAAGTTTTACTCTTAAGGTTCTGAAAGATTTAATAGGAACAGTATGTGGTCCTCCATA
GGATGAATTGGTTGCAATTGAGCAATAGTGTCAAAATCATCAGTTGATCAATTCTTCTGC
ATAACCATTTATGTGAAATTGACTAGAAAACAAGTTGCAAGAGAAAAATAAAGCTTTCTG
GTTTAGCTTGTTGTGTTAGCCCTTTCAGACACAGGTCAGTGTTGAACATATTTCTAAGAT
AATTAGGTTAGCTAAGATGAGAGGCAAACTCTATTTATTTGTGACCCTAAAAATGGTAGA
CTTACAAACGCCTAACTTAATCATACTCAATCTTCATGTCTACTTCAGTTAAAGAGAATA
CAATTACAACAAGTACCCAACCCGCAATCACCAATAAAAACTAAACAATCTAACAGAGAT
ATTGTTTACACTAAAGAAACAAAAACATTAAGTAAATTGACCAATGACTCCCATCGTACT
ACTGTCG
>lcl|Seq158084 No definition line found
TCAGCCGCGCAGGTATCTTCTACTACAGTGATGACATATCATTTCCATGTCTTGCAGACT
CTCTCGCATATACTGACAGTGATATCACTTCCTACTAGCTAGCTGCTACTTGAAACTAAA
GTTTTACTCTTAAGGTTCTGAAAGATTTAATAGGAACAGTATGTGATTCTCCATAGGATG
AATTGGTTGCTAATTGAGTCAAATAGTGTCAAAATCAATCAGTTGATCATTCTTCTGCAT
AACCATTTATGTAGAAATTGACTAAGAAAACAAGTTGCAAGAAGAAAAATAAAGACTTTT
ACTGGTTTAAGCTTTGTTAGTGTTAGCCCTTTACAAGACACAGGTCTAGTGTTGAACATA
TTTACTAAGATAATTAGGTTAGACTAAAGATGAGTAGCAAACTCTATTTATTGTGTACCC
AAAAATGGTAGACTTACAAACGTACCCTAATTAATCATACTCATTCTTCATGTCCTACTT
ACAAGGTTAAAGAAGTAATAACAATTACAAACTAAACGTAACCTAACCCGACAACTACAA
CCAATAAAAACTAAAAC

[piroyon]

Re: tell me the different between two version cap?

I have a same problem.

"A version of CAP3 for a 64-bit Linux system with an Intel processor" has a new doc.

However, I don’t know how to get a contig from your sequences using a new version cap3...

$ diff old_doc new_doc

204a208,211
> If the option -k 0 is given, then no read end is clipped and
> the whole read is used in assembly. Otherwise, the following procedure
> is used to determine and clip poor read ends.
>
349a357,391
> Short reads
>
> The default values for some of the parameters are selected for
> assembly of regular reads of lengths 500 to 1000 bp.
> For assembly of short reads of lengths 20 bp, the following options
> should be used to change the values for those parameters accordingly.
>
> -i 30 -j 31 -o 18 -s 300
>
> Note that using short reads increases the likelihood of producing assemblies
> with false joins. Below we explain the options for short reads.
> Overlaps between reads are quickly computed by finding segment pairs
> (ungapped alignments) and combining segment pairs into chains.
> The -i option is used to specify a score cutoff on segment pairs.
> The score of a segment pair with 19 base matches and 1 base mismatch
> is 2 * 19 + (-5) * 1 = 33, where each base match is given a score of 2
> and each mismatch is given a score of -5.
> The -j option is used to specify a score cutoff on chains of segment pairs,
> where the score of a chain is the sum of scores of each segment pair
> minus penalties for gaps between segment pairs.
> The score of a chain consisting of one segment pair is simply the score of
> the segment pair.
>
> After a high scoring chain of segment pairs between two reads is computed,
> an overlap between the reads is computed as an optimal local alignment
> between the reads, where the chain is used to limit the
> computation to a small area of the dynamic programming matrix.
> Unlike the scores of segment pairs and chains, the score of an overlap
> is weighted by base quality values. Thus, an overlap
> with 19 base matches, 1 base mismatch, and 0 gap has a score
> of 10 * [2 * 19 + (-5) * 1] = 330, assuming that each base
> has a quality value of 10.
> The -o option is used to specify a length cutoff on overlaps,
> whereas the -s option is used to specify a score cutoff on overlaps.
>
409a452
> Sanzhen Liu and Pat Schnable for suggesting the options -i, -j, -k.

[piroyon]

Hi, your sequences could be assembled:

$cap3 tmp.fas -y 250 -p 80 -m 4

The default "-m" value had been changed?
I don't know it because I've never see the new source cap3.c.

[piroyon]

Sorry, the -m value of old cap3 is actually 2.
They have other complex causes...