Reference Manual

AGP files provided with the --agp argument (see Section 3.1.2 below) must conform the NCBI’s AGP v2.0 specification. Please visit the following link for details:

NCBI AGP v2.0 Specification

GAEMR relies on Biopython for most FASTA processing, so input FASTA sequences should conform to the format in the following link:

FASTA Sequence Format

GAEMR requires a list of BAM files in CSV file format supplied to it view the --read_list (-l). The first line of the file is the header and should look like this:

#name,lib_type,mean_read_len,dir,insert_size,files

The following table describes the header items. Note that some of these options do not apply to certain library types, and therefore should be left blank within the CSV file (see example below):

[table delimiter=”|”]

Field|Description

name| The library’s identifier

lib_type| The library type (fragment, jump, unpaired)

mean_read_length| The mean read length of the reads

dir| The expected direction of paired reads with respect to each other (FR is forward and reverse, RF is reverse and forward)

insert_size| The mean insert size for the library

files| The full file path for the library’s bam file

[/table]

The following example demonstrates a read list file specifying fragment pair, jump pair, and PacBio unpaired data, in that order.

#name,lib_type,mean_read_len,dir,insert_size,files
C0DJUACXX.1.Pond-121053,fragment,101,fr,200,/my/bam/file/directory/fragments.bam
C0DJUACXX.2.Solexa-76388,jump,101,rf,3000,/my/bam/file/directory/jumps.bam
All.PacBio,unpaired,900,,,/my/bam/file/directory/long_reads.bam

Note that ‘dir’ and ‘insert_size’ are not supplied for the PacBio data because they do not apply to this read type.

 

At least one of the following arguments must be specified for GAEMR to run:

[table delimiter=”|”]
Argument|Description|Example
-s SCAFFOLDS~~--scaffolds=SCAFFOLDS|Specifies a scaffolds FASTA file to use for analysis (Default=none)|GAEMR.py -s scaffolds.FASTA
-c CONTIGS~~--contigs=CONTIGS|Specifies a contigs FASTA file to use for analysis (Default=none)|GAEMR.py -c contigs.FASTA
[/table]

The following arguments may be used in any combination to produce more data or to further customize the analysis run.

[table delimiter=”|” nl=”~~”]
Argument|Description|Example
-h~~--help|Prints GAEMR usage information to screen|GAEMR.py -h~~GAEMR.py --help
-r REFERENCE~~--reference=REFERENCE|Specifies the reference file to use for analysis (default=none)|GAEMR.py -r reference.FASTA~~GAEMR.py --reference=reference.FASTA
-a AGP~~--agp=AGP|Specifies an AGP file for analysis(default=none)| GAEMR.py -a genome.agp~~GAEMR.py --agp=genome.agp
-l READ_LIST_FILE~~--readlist=READ_LIST_FILE|Specifies the list of reads to use for alignment(default=none)|GAEMR.py -l reads.list~~GAEMR.py --readlist=reads.list
-t THREADS~~--threads=THREADS|Indicates the number of threads to use for multi-threadable processes(default=1)| GAEMR.py -t 4~~GAEMR.py --threads=4
-w WINDOW_SIZE~~--window_size=WINDOW_SIZE|Selects the window size for chart outputs (default=1000)|GAEMR.py -w 500~~GAEMR.py --window_size=500
-n ASSEMBLY_NAME~~--assembly_name=ASSEMBLY_NAME|Name of assembly (default=assembly)|GAEMR.py -n my_assembly~~GAEMR.py --assembly_name=my_assembly
-m ASSEMBLER~~--assembler=ASSEMBLER|Assembler used to generate consensus(default=assembler)|GAEMR.py -m allPaths~~GAEMR.py --assembler=allPaths
-o OUTPUT~~--output=OUTPUT|Output header (default=assembly)|GAEMR.py -o my_assembly~~GAEMR.py --output=my_assembly
-b BLAST_TASK~~--blast_task=BLAST_TASK|Type of blast to run for contamination check: blastn, dc-megablast, megablast (default=megablast)|GAEMR.py -b blastn~~GAEMR.py --blast_task=blastn
-i ALIGNER~~--aligner=ALIGNER|Aligner to use for reads alignments: bwa or bowtie2(default=bwa)|GAEMR.py -i bowtie2~~GAEMR.py --aligner=bowtie2
-k KMER_SIZE~~--kmer_size=KMER_SIZE|k-mer size for repeat and k-mer coverage stats (default=29)|GAEMR.py -k 100~~GAEMR.py --kmer_size=100
--no_blast|Do not run blast jobs (default=false)|GAEMR.py --no_blast
--minScaffSize=MINSCAFFSIZE|Minimum scaffolds size for inclusion in analysis(default=1)|GAEMR.py --minScaffSize=1000
--minConSize=MINCONSIZE|Minimum contig size for inclusion in analysis(default=1)|GAEMR.py --minConSize=500
--minGapSize=MINGAPSIZE|Minimum length of a string of Ns to determine a gap(default=10)|GAEMR.py --minGapSize=50
[/table]

 

Table 1.1: Basic Assembly Stats (basic_assembly_stays.py)

 

 

The header for this table identifies the name of the assembly, as specified by the  --assembly_name (-n) option. The rows are described by the following table:

[table delimiter=”|”]

Descriptor| Description

Assembler|The assembler used for assembly, as specified by the --assembler(-m) option

Contigs|The number of contigs in the assembly

Max Contig|The length(in bases) of the largest contig in the assembly

Mean Contig|The mean size of the contigs in the assembly

Contig N50| The length of the contig in which all larger contigs in the assembly would compose 50% of the genome

Contig N90| The length of the contig in which all larger contigs in the assembly would compose 90% of the genome

Total Contig Length| The sum of the lengths of all contigs in the assembly

Assembly GC|The percent of G and C bases in the assembly

Scaffolds|The number of scaffolds in the assembly

Max Scaffold|The length of the largest scaffold in the assembly

Mean Scaffold|The mean size of the scaffolds in the assembly

Scaffold N50|The length of the scaffold in which all larger scaffolds in the assembly would compose 50% of the genome

Scaffold N90| The length of the scaffold in which all larger scaffolds in the assembly would compose 90% of the genome

Total Scaffold Length|The sum of the lengths of all scaffolds in the assembly

Captured Gaps| The number of gaps spanned by jump reads

Max Gap|The predicted size of the largest gap

Mean Gap|The predicted mean size of all the gaps

Gap N50| The size of the gap in which all larger gaps in the assembly would compose 50% of all missing sequence

Total Gap Length|The sum of the sizes of all gaps in the assembly

[/table]

 

Table 1.1: rRNA Analysis Summary (analyze_rna_hits.py)

 

This table provides results on GAEMR’s rRNA analysis module.  The contents are described in the following table:

[table delimiter=”|”]

Descriptor| Description

Gene| The gene identified in the assembly

Total Copies| The number of copies of the gene in the assembly

Lineage| The lineage of the organism found

Number of Organisms Found| The number of different organisms identified by rRNA analysis

Organism IDs| The genus of the organism

[/table]

Figure 1.3: Cumulative Sizes (basic_assembly_stats.py)

 

The Cumulative Sizes plots give the user an indication of how many contigs and scaffolds compose 50% and 90% of the analyzed genome. In this example, the contigs plot (top) demonstrates that 50% of this genome is in 10 contigs, and 90% in 32 contigs. The scaffold plot (bottom) shows 50% of the genome is in 2 scaffolds, and 90% of the genome is in 6 scaffolds.

 Figure 2.1: Coverage Line Plot and GC (generate_bam_plots.py)

 

This two-part figure displays GC content, fragment read coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation., and jump read coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation. across the genome as represented by the x-axis. Black vertical lines represent scaffold breaks, and tick-marks represent 100KB intervals. The red line in the upper box shows variation in GC content. Green and blue lines in the bottom box represent fragment and jump read coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation., respectively. In this example, we can see that GC content for this genome is generally about 55%, with multiple short dips to 20% or lower. The assembly has 200x fragment coverage, and 75x jump coverage. Note that we can also see a drop in fragment and jump coverage in Scaffold00006 at 100KB. Locations where coverage drops to zero are likely gaps in the scaffold sequence.

Figure 2.2:Scaffold GC Coverage Histogram (generate_bam_plots.py)

The top and bottom histograms provide information on jump and fragment coverage in the genome. Since each base can not be presented in a histogram, bins of coverage are used. The y-axis indicates the frequency of the bins, and the x-axis the coverage. For example, in the fragments histogram, we can see that the most frequent coverage in the genome is 200x coverage.

Table 2.3: Simple BAM Stats(get_simple_bam_stats.py)

 

The module that creates this table generates a column for each library type in the assembly. The table below describes the contents of the table:

[table delimiter=”|”]

Descriptor| Description

Total Reads| The number of reads in the assembly for each library type

Paired Reads| The number of paired reads for each library type

Duplicates| The number of duplicate reads for each library type

Total Read 1| The total number of forward reads for each library type

Total Read 2| The total number of reverse reads for each library type

Mapped| The number and percentage of reads mapping to the reference

Singletons|The number and percentage of reads in which only one mate of a read pair is present in the assembly

Mapped w/ Mate| The number and percentage of mapped reads of which the mated read is also present in the assembly

Properly Paired| The number and percentage of mated reads that are properly oriented with respect to each other and are the expected distance from each other based on insert size information

Cross-chromosome| The number and percentage of paired reads in which mates assembled in different chromosomes

Cross-chromosome (MQ >= 5)| The number and percentage of paired reads in which one mate maps to another chromosome with a mapping quality greater than or equal to 5

[/table]

Table 2.4: Sequence Coverage Stats (get_bam_coverage_stats.py)

 

The module that creates this table generates a column for each library type in the assembly. The table below describes the contents of the table:

[table delimiter=”|”]

Descriptor| Description

Average Cvg| The mean sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. across the assembly

Cvg Std Dev| The deviation in sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. from the mean for the majority of reads in the assembly.

Median Cvg| The median sequence depth across the assembly

Cvg Mode| The most frequent sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. level in the assembly

[/table]

Table 2.5: Physical Coverage Stats (get_bam_coverage_stats.py)

 

The module that creates this table generates a column for each library type in the assembly. The table below describes the contents of the Physical CoverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation. Stats table:

[table delimiter=”|”]

Descriptor| Description

Average Cvg| The mean physical coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation. across the assembly

Cvg Std Dev| The deviation in physical coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation. from the mean for the majority of reads in the assembly.

Median Cvg| The median physical coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation. across the assembly

Cvg Mode| The most frequent physical coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation. level in the assembly

[/table]

Table 2.6: Insert Size Stats (plot_insert_size.py)

 

The table presents insert size statistics for fragment (top half) and jump (bottom half) reads. This example genome has two jump libraries, hence two columns of insert size data.

[table delimiter=”|”]

Descriptor| Description

Pair Orientation| The orientation of the left and right reads of a pair with respect to each other(F=Forward, R=Reverse)

Read Pairs| The number of read pairs in the genome assembly

Mean Insert Size| The average insert size for the read pair

Median Absolute Deviation| A measure of the magnitude of dispersion of the insert sizes

Min Insert Size| The smallest insert size in the pool of read pairs

Max Insert Size| The largest insert size in the pool of read pairs

Width of 50 Percent|  The “width” of the bins, centered around the median, that encompass 50% of all read pairs

With of 90 Percent| The “width” of the bins, centered around the median, that encompass 90% of all read pairs

Sample|  The sample ID as specified by the BAM header

Library|The library ID as specified by the BAM header

Read Group| The read group as specified by the BAM header

[/table]

 

Figure 2.7: Scaffold Insert Size (plot_insert_size.py)

This plot shows a histogram of jump read insert sizes with an associated box plot. The box plot (using the histogram x-axis) presents several insert size metrics: minimum, first quartile, second quartile, median, third quartile, maximum, and outliers. The minimum insert size in this example is 1, and the first quartile, represented by the left-most dashed line, is the first quartile (25th percentile). The box represents the second quartile (from ~1250bp to ~2300bp), all reads between the 25th and 75th percentile (third quartile). The line bisecting the box is the median (1713bp). The horizontal dashed line to the right of the box represents all insert sizes above the 75th percentile. The vertical dash, or ‘upper fence’, represents the maximum value for the insert size, and the plus signs beyond the fence are outliers.

The insert size histogram presents the same data as the jump insert size histogram but for fragment reads. Note that in addition to an upper fence, the lower fence is displayed at approximately 130bp.

Figure 2.8: Coverage Anomalies Scatter Plot (IdentifyCoverageAnomalies.py)

The Coverage Anomalies Scatter Plot shows the points along the genome where coverage is greater than three standard deviations from the norm. For example, this plot shows positions of high coverage at approximately 600 KB and 2.1 MB(~120x coverage).

 

Figure 2.9: Scaffold Coverage Anomalies Cluster Length (IdentifyCoverageAnomalies.py)

 

This bar graph represents clusters of regions with anomalous coverage. In this example, four anomalous clusters are represented, with  the largest cluster group (#2) containing 27 positions with anomalous coverage.

 

Figure 2.10: Scaffold Coverage Anomalies Change in Coverage Scatter Plot (IdentifyCoverageAnomalies.py)

 

This scatter plot represents change in coverage from one base position to the next across the genome. For example, there is an approximately 235x change in coverage near 800 KB in this genome, as shown by the dot in the upper left quadrant of the plot.

 

Figure 2.11: Scaffold Coverage Anomalies Change in Coverage Histogram (IdentifyCoverageAnomalies.py)

 

This histogram gives the analyst an idea of which coverage levels are most prevalent in the genome. Since a base-by-base analysis is not feasible sequence windows, or bins,  are used. In this figure, for example, their are over 700 sequence windows in which the mean coverage change from one window to the next is about 65x. 

 

Table 3.1: Kmer Copy Number (kmer_copy_number.py)

 

This table presents the frequency of Kmer copy numbers, with the left-most column serving as the header and each row containing frequency data for the copy number indicated by the header.

[table delimiter=”|”]

Descriptor| Description

Count| The total number of Kmers for the given copy number.

BasePct| The percentage of bases from the genome that have the given copy number.

KmerPct| The percentage of Kmers that have the given copy number.

[/table]

 

Table 4.1: Gap End Analysis Table (analyze_gap_ends.py)

 

 

 

This table, as the header indicates, presents metrics for both captured gaps and uncaptured ends.

[table delimiter=”|”]

Descriptor| Description

Number| The total number of uncaptured ends and captured gaps (uncaptured ends should always be twice the scaffold count, and captured gaps should be the number of contigs minus scaffolds)

Average Complexity| The percentage of 5-mers near gap ends that occur once in the genome

Less than 75% Complexity| The percent of gaps that are less than 75% complex.

Average GC| The percent of Gs and Cs in the gap flanking sequence

Less than 30% GC| The number of gaps composed of less than 30% Gs and Cs

Greater than 70% GC| The number of gaps composed of greater than 70% Gs and Cs

Average Copy Number| The average number of copies of the gap sequence in the genome

[/table]

 

Figure 4.2: CG Sizes (analyze_gap_ends.py)

 

This histogram displays the percentage of gaps of various lengths. For example, 10% of the gaps in the analyzed genome are below 50 bases wide, and 4% are between 600 and 650 bases wide.

 

Figure 4.3: CG Complexity (analyze_gap_ends.py)

 

The gap flank complexity graph gives an indication of repeat content in the sequence flanking the gaps in the analyzed genome. It is calculated by examining all 5-mers in a gap’s flanking sequence (fifty bases on each side) and assessing the percentage of 5-mers that occur only once.  In this example, eleven gaps have a 5-mer composition in which 98% of 5-mers occur only once.

 

Figure 4.4: CG Copy Number (analyze_gap_ends.py)

 

The graph presents the range of copy numbers for sequence flanking gaps as a percentage of all gaps. For example, the figure above shows that 2% of all gap flank sequences have a copy number of five.

Figure 4.5: CG GC (analyze_gap_ends.py)

This histogram shows the frequency (in percent) of Gs and Cs across a range of GC percentages for captured gap flanking sequences. For example, 1% of the gaps have flanking sequence with 60% GC content in the figure above.

 

Figure 4.6: UE Distinctness (analyze_gap_ends.py)

 

The uncaptured end flank complexity graph gives an indication of repeat content in the sequence at the end of a scaffold in the analyzed genome. It is calculated by examining all 5-mers in the end sequence (fifty bases) and assessing the percentage of 5-mers that occur only once.  In this example, six gaps have a 5-mer composition in which 96% of 5-mers occur only once.

 

Figure 4.7: UE Copy Number (analyze_gap_ends.py)

 

Like Figure 4.3, this graph displays sequence copy number, but in this case as a percentage of uncaptured ends. In this example, 4% of the uncaptured ends have sequence with five copies within the genome.

 

Figure 4.8: UE GC (analyze_gap_ends.py)

 

This histogram shows the frequency (in percent) of Gs and Cs across a range of GC percentages for uncaptured ends. For example, 2% of the gaps have uncaptured end sequence with 34% GC content in the figure above.

 

Figure 5.1: Alignment to Reference (run_nucmer.py)

 

This dotplot presents a 1-to-1 alignment between the assembled genome  and the reference sequence, with reference  and coordinates drawn on the x-axis and scaffolds and their IDs on the y-axis. Red lines and dots represent sequence that aligns in the same sense as the reference, while blue lines (none shown here) and dots represent sequence that is complemented with respect to the reference. Note that the scaffolds marked with an asterisk are actually complemented in the assembly FASTA with respect to the reference, but are drawn in the same sense to show the actual order and orientation of the genome.

Table 5.2: Scaffold Accuracy (run_scaffold_accuracy.py)

 

 

This table provides various statistics relevant to how accurately the genome assembled. Note that this actual mate pairs from the libraries are not used, but rather ‘pairoids’ are generated from the reference sequence (sequences with a fixed number of bases between them in the reference sequence designate a pairoid) and then mapped to the assembly FASTA, and various statistics are collected to give a general indication of how well the genome assembled with respect to the reference. Raw numbers and percentages are provided in the second and third columns.

Descriptor| Description
Total Input Pairs| The number of pairoids sampled.
Multiply Mapped| The number and percent of pairoids from the sample that map to more than one location in the genome's reference sequence
Unaligned| The number and percent of pairoids that do not map to the genome's reference sequence
Cross Chromosome|  The number and percent of pairoids in which pairs map to two different chromosomes
Invalid Orientation| The number and percent of pairoids  in which the orientation is invalid. For most libraries, valid orientation is when the left read has a forward orientation, and the right read has a reverse orientation
Invalid Length| The number and percent of pairoids separated by more or less than the allowable number of bases apart
Valid | The number and percent of pairoids with valid orientation and length
Valid Circular | The number and percent of pairoids that create circularity from one end of the scaffold to the other.

Note that the definition of ‘cross chromosome’ depends on the reference used for the assembly. Specifically, when the reads of a pairoid map to two different elements (chromosomes, plasmids scaffolds)  in the FASTA file, they will be labeled as cross chromosome. In a finished reference with multiple chromosomes, the meaning is as expected. If a finished reference contains chromosomes and plasmid, it could mean that one read in a pairoid maps to a plasmid, and the other to a chromosome, and in an unfinished reference, it could mean that one read in a pairoid maps to a different scaffold that may or may not be part of the same chromosome.

Table 5.3: Kmer Coverage (kmer_coverage.py)

 

This table provides more detailed information on the kmer coverage across the analyzed genome by providing the total number bases in the kmers, bases covered in the reference, and percent of reference covered for multiple categories.

Descriptor| Description
Covered | The total bases in the kmers, the total bases that align to the reference, and percentage of total that align to the reference
DistinctCovered | The distinct bases in the kmers with copy number of one, the total bases that align to the reference, and percentage of total that align to the reference
OverCovered | The bases in the kmers with copy number greater than one, the total bases that align to the reference, and percentage of total that align to the reference
Novel | The bases in kmers that are novel and the percentage of the genome that these novel bases composes
NovelDistinct |  The bases in kmers with a copy number of one that are novel and the percentage of the genome that these novel bases composes

Figure 5.4: Coverage and GC Line Plot 

Like figure 2.1, this plot displays GC Content, fragment read coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation., and jump read coverageThe number of reads and read templates covering a given base position in a genome assembly. The unsequenced regions between read pairs is included in this calculation., but in this plot, the x-axis represents the reference genome and not the actual assembly.

 

Figure 5.5: GC Coverage Histogram (generate_bam_plots.py)

The histograms shown above display the frequency of various coverage levels when jump and fragment reads are aligned to the reference genome. In this example, we can see that the most frequent coverage is about 70x to 75x in jump reads, and about 210x to 220x in fragment reads.

Table 5.6: Reference Simple BAM Stats Table (get_simple_bam_stats.py)

Like Table 2.1, this table presents simple BAM stats, but with respect to the reads as they align to the reference sequence.

Descriptor| Description
Total Reads| The number of reads in the assembly for each library type
Paired Reads| The number of paired reads for each library type
Duplicates| The number of duplicate reads for each library type
Total Read 1|  The total number of forward reads for each library type
Total Read 2|  The total number of reverse reads for each library type
Mapped| The number and percentage of reads mapping to the reference
Singletons|The number and percentage of reads in which only one mate of a read pair is present in the assembly
Mapped w/ Mate| The number and percentage of mapped reads of which the mated read is also present in the assembly
Properly Paired| The number and percentage of mated reads that are properly oriented with respect to each other and are the expected distance from each other based on insert size information
Cross-chromosome| The number and percentage of paired reads in which mates assembled in different chromosomes
Cross-chromosome (MQ	gt;= 5)| The number and percentage of paired reads with mapping quality greater than or equal to 5 in which mates assembled in different chromosomes

Table 5.7: Reference Sequence Coverage Stat (get_bam_coverage_stats.py)

This table provides statistics on how well the fragment and jump read sequences cover the reference sequence.

Descriptor	Description
Average Cvg	The mean sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. across the reference
Cvg Std Dev	The deviation in sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. from the mean for the majority of reads in the reference.
Median Cvg	The median sequence depth across the reference
Cvg Mode	The most frequent sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. level in the reference

 

Table 5.8: Physical Coverage Stats (get_bam_coverage_stats.py)

 

 

This table provides statistics on how well the fragment and jump reads physically cover the reference sequence.

Descriptor	Description
Average Cvg	The mean sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. across the reference
Cvg Std Dev	The deviation in sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. from the mean for the majority of reads in the reference.
Median Cvg	The median sequence depth across the reference
Cvg Mode	The most frequent sequence coverageThe number of sequenced read bases at a given position in a genome assembly. This does not include the unsequenced portions of a read template. level in the reference

 

Figure 5.9: Insert Size (plot_insert_size.py)

 

This insert size histogram presents the same data as the jump insert size histograms shown in figures 5.8 and 2.5, but for fragment reads. Like figure 2.5, this example shows a lower fence (at ~140bp) as well as an upper fence (at ~ 250bp).

This plot shows a histogram of jump read insert sizes with an associated box plot. The box plot (using the histogram x-axis) presents several insert size metrics: minimum, first quartile, second quartile, median, third quartile, maximum, and outliers. The minimum insert size in this example is 1, and the first quartile, represented by the left-most dashed line, is the first quartile (25th percentile). The box represents the second quartile (from ~1250bp to ~2300bp), all reads between the 25th and 75th percentile (third quartile). The line bisecting the box is the median (1730 bp). The horizontal dashed line to the right of the box represents all insert sizes above the 75th percentile. The vertical dash, or ‘upper fence’, represents the maximum value for the insert size, and the plus signs beyond the fence are outliers.

Table 5.10: Insert Size Stats (plot_insert_size.py)

 

The table presents insert size statistics for fragment (top half) and jump (bottom half) reads. This example genome has two jump libraries, hence two columns of insert size data.

Descriptor	Description
Pair Orientation	The orientation of the left and right reads of a pair with respect to each other(F=Forward, R=Reverse)
Read Pairs	The number of read pairs in the genome assembly
Mean Insert Size	The average insert size for the read pair
Median Absolute Deviation	A measure of the magnitude of dispersion of the insert sizes
Min Insert Size	The smallest insert size in the pool of read pairs
Max Insert Size	The largest insert size in the pool of read pairs
Width of 50 Percent	The "width" of the bins, centered around the median, that encompass 50% of all read pairs
With of 90 Percent	The "width" of the bins, centered around the median, that encompass 90% of all read pairs
Sample	The sample name, if available in the BAM file header
Library	The library name, if available in the BAM file header
Read Group	The read group name, if available in the BAM file header

Figure 5.11: Comparison To Reference 

The Comparison to Reference Table shows a base-count comparison of shared and novel sequence between the assembled genome and the reference genome. In this example, there are 39 sequence regions that are present in the reference but not the assembly, and 44 sequences present in the assembly but not the reference.

Descriptor	Description
Fasta File ID	The names of the fasta files being compared
Total Length (bp)	The total length of the reference and assembly sequences
Total Novel Regions	The total number of novel regions in the reference and assembly sequences
Total Novel Bases (bp)	The total number of bases among all novel regions
Average Novel Region Size (bp)	The average size of novel regions in each sequence
Largest Novel Region Size (bp)	The largest novel region in each sequence
N50 Novel Region Size (bp)	The N50 size of the novel regions in each sequence
Pct Covered	The percent of sequence covered in one sequence by the comparison sequence
Pct Identity	The percent identity, or sequence similarity, of the two sequences to each other.

 

Figure 5.12: Coverage Anomalies Scatter Plot (IdentifyCoverageAnomalies.py)

The Coverage Anomalies Scatter Plot shows the points along the genome where coverage is greater than two standard deviations from the norm.

Figure 5.13: Coverage Anomalies Bar Plot (IdentifyCoverageAnomalies.py)

 

This bar graph represents cluster of regions with anomalous coverage. In this example, ten anomalous clusters are represented, with the largest cluster group (#4) containing 8 positions with anomalous coverage.

Figure 5.14: Coverage Change Across Bases Scatter Plot (IdentifyCoverageAnomalies.py)

 

This scatter plot represents change in coverage from one base position to the next across the genome. For example, there is an approximately 105x change in coverage near the beginning of this genome, as shown by upper-left-most dot in the figure.

Figure 5.15: Coverage Change Across Bases Histogram (IdentifyCoverageAnomalies.py)

 

This histogram shows the frequency of coverage changes and gives an indication of how variable the coverage is in the genome.  In this exmaple, most coverage changes are near ~70x. 

 

Figure 6.1: Blast Bubbles (blast_bubbles.py)

 

This bubble plot shows four pieces of information to display the taxonomic makeup of the genome assembly. Coverage depth is represented on the y-axis, and percent GC on the x-axis.The color key beneath x-axis  indicates the different organisms in the assembly based on Blast results, and the number inside brackets indicates how many contigs match that genus. Contigs are drawn as bubbles at the intersection of their coverage depth and percent GC values. The size of the bubble indicates the length of the contig, and the color its taxonomy. In this example, we see that there is a small contig with just under 50x coverage and ~31% GC content that aligns to Escherichia, and that most of the large contigs in this assembly range from 200x to 300x coverage, 47% to 57% GC, and that all contigs align to Escherichia genus.

Figure 6.2: Blast Heat Map (blast_map.py)

 

The Blast Heat Map indicates the degree to which the sequence in the contigs of the assembly match the reference sequence, with darker shading indicating a strong match, and lighter shading indicating a weak match. In this example we can see that all the contigs strongly match Escherichia, though contigs 36 and 24 have each have a small region with weaker matches.

Table 6.3: Blast Taxonomy Details (get_blast_hit_taxonomy.py)

This table provides details taxonomic information for each contig in the genome, as described below:

[table delimiter=”|”]

Descriptor| Description

QueryID | The contig name

QueryLen | The length in bases of the contig

QueryHitLen | The length in based of the matched alignment

PctCovered | The percent of the contig matched by the aligned sequence

TaxonomicString | The full taxonomic information, from domain to species, of the aligned sequence

[/table]

In this table, we can see that 77%, or 128 of 166kb, of contig000006 aligns to Escherichia coli BL21(DE3).

 

Table 7.1: rRNA Analysis Details (analyze_rna_hits.py)

This table displays 16s rRNA hits

[table delimiter=”|”]

Descriptor| Description

Gene | The aligned gene

Genus | The genus of the organism of the aligned gene

Query ID | The contig containing the aligned gene sequence

Query Start | The start position of the aligned gene sequence

Query End | The end position of the aligned gene sequence

Hit Direction | The direction of the alignment with regards to the contig sequence

Length | The length in bases of the sequence alignment

Taxonomy | The taxomomic information of the aligned sequence

[/table]

Table 8.1: Contig Details (make_detailed_assembly_table.py )

 

 

The table above provides summary details for each contig in the assembly as listed below:

 

[table delimiter=”|”]

Descriptor| Description

Contig| The name of the contig

Scaffold| The scaffold that contains the contig

Length| The length of the contig

GC| The average GC content of the contig

Coverage (F/J/L)| The mean coverage for the contig. A breakdown of fragment(F), jump(J), and long read(LR) coverage is shown in the parentheses

BLAST Hit| The genus of the top blast hit for the contig

BLAST Covered| The percent of the contig’s sequence aligning to the top blast hit

[/table]

 

Table 8.2: Scaffold Details (make_detailed_assembly_table.py )

 

 

The Scaffold Details Table provides information on the number of contigs, length, %GC content, coverage, BLAST alignment, and circularity for each scaffold in the assembly.

[table delimiter=”|”]

Descriptor| Description

Scaffold| The name of the scaffold

Num Contigs| The number of contigs contained within the scaffold

Length| The length of the scaffold

GC| The average GC content of the scaffold

Coverage (F/J/L)| The mean coverage for the scaffold. A breakdown of fragment(F), jump(J), and long read(L) coverage is shown in the parentheses

BLAST Hit| The genus of the top blast hit for the scaffold

BLAST Covered| The percent of the scaffold’s sequence aligning to the top blast hit

Circular| The number of circular reads in the scaffold

[/table]