CLI Internals

The following page describes the internal API used by the Command Line Pipeline. These functions and objects are not meant for interactive usage. So this page is useful if you want to change the behavior of the molecule counting pipeline.

velocyto.counter module

velocyto.transcript_model module

class velocyto.transcript_model.TranscriptModel(trid: str, trname: str, geneid: str, genename: str, chromstrand: str)

Bases: object

A simple object representing a transcript model as a list of vcy.Feature objects

trid

trname

geneid

genename

chromstrand

list_features

property start: int

This should be accessed only after the creation of the transcript model is finished (i.e.) after append_exon has been called to add all the exons/introns

Type

NOTE

property end: int

This should be accessed only after the creation of the transcript model is finished (i.e.) after append_exon has been called to add all the exons/introns

Type

NOTE

ends_upstream_of(read: velocyto.read.Read) → bool

intersects(segment: Tuple[int, int], minimum_flanking: int = 5) → bool

append_exon(exon_feature: velocyto.feature.Feature) → None

Append an exon and create an intron when needed

Parameters

exon_feature (vcy.Feature) – A feature object represneting an exon to add to the transcript model.

chop_if_long_intron(maxlen: int = 1000000) → None

Modify a Transcript model choppin the 5’ region upstram of a very long intron To avoid that extremelly long intron mask the counting of interal genes

Parameters

maxlen (int, default=vcy.LONGEST_INTRON_ALLOWED) – transcript model tha contain one or more intronic interval of len == maxlen will be chopped

Returns

• Nothing it will call _remove_upstream_of or _remove_downstream_of on the transcript model

• its name will be changed appending _mod to both trid and trname

velocyto.segment_match module

class velocyto.segment_match.SegmentMatch(segment: Tuple[int, int], feature: velocyto.feature.Feature, is_spliced: bool = False)

Bases: object

segment

feature

is_spliced

property maps_to_intron: bool

property maps_to_exon: bool

property skip_makes_sense: bool

If the SKIP in the segment matches some extremity of the feature and therefore can be interpreted as a splice event

velocyto.feature module

class velocyto.feature.Feature(start: int, end: int, kind: int, exin_no: str, transcript_model: Optional[Any] = None)

Bases: object

A simple class representing an annotated genomic feature (e.g. exon, intron, masked repeat)

start

end

transcript_model

kind

exin_no

is_validated

property is_last_3prime: bool

get_downstream_exon() → Any

To use only for introns. Returns the vcy.Feature corresponding to the neighbour exon downstream

Note

In a 15 exons transcript model: Downstream to intron10 is exon11 or the interval with index 20 if strand “+”. Downtream to intron10 is exon10 or the interval with index 10 if strand “-“

get_upstream_exon() → Any

To use only for introns. Returns the vcy.Feature corresponding to the neighbour exon downstream

Note

In a 15 exons transcript model: Upstream to intron10 is exon9 or the interval with inxex 18 if strand “+”. Upstream to intron10 is exon11 or the interval with inxex 8 if strand “-“

ends_upstream_of(read: velocyto.read.Read) → bool

The following situation happens

Read

*|||segment|||-?-||segment|||????????

???????|||||Ivl|||||||||*

doesnt_start_after(segment: Tuple[int, int]) → bool

One of the following situation happens

*||||||segment|||||????????

||||Ivl|||||

*|||||||||||||Ivl||||||||||????????????

*|||||||||||||Ivl||||||||||????????????

*|||||||||||||Ivl||||||||||????????????

intersects(segment: Tuple[int, int], minimum_flanking: int = 5) → bool

contains(segment: Tuple[int, int], minimum_flanking: int = 5) → bool

One of following situation happens

—–||||||segment|||||—–

|||||||||||||Ivl||||||||||||||||

—–||||||segment|||||—–

|||||||||||||Ivl||||||||||||||||

—–||||||segment|||||—–

|||||||||||||Ivl||||||||||||||||

where — idicates the minimum flanking

start_overlaps_with_part_of(segment: Tuple[int, int], minimum_flanking: int = 5) → bool

The following situation happens

—|||segment||—

|||||||||||||Ivl||||||||||||||||

where — idicates the minimum flanking

end_overlaps_with_part_of(segment: Tuple[int, int], minimum_flanking: int = 5) → bool

The following situation happens

—|||segment||—

|||||||||||||Ivl||||||||||||||||

where — idicates the minimum flanking

velocyto.indexes module

class velocyto.indexes.TransciptsIndex(trascript_models: List[velocyto.transcript_model.TranscriptModel])

Bases: object

transcipt_models

tidx

maxtidx

property scan_not_terminated: bool

Return false when all the chromosome has been scanned

find_overlapping_trascript_models(read: velocyto.read.Read) → Set[velocyto.transcript_model.TranscriptModel]

Finds all the Transcript models the Read overlaps with

Parameters

read (vcy.Read) – the read object to be analyzed

Returns

matched_transcripts – TranscriptModel the read is overlapping with and values the kind of overlapping it is one of vcy.MATCH_INSIDE (1), vcy.MATCH_OVER5END (2), vcy.MATCH_OVER3END (4)

Return type

set of vcy.TranscriptModel

class velocyto.indexes.FeatureIndex(ivls: List[velocyto.feature.Feature] = [])

Bases: object

Search help class used to find the intervals that a read is spanning

property last_interval_not_reached: bool

reset() → None

It set the current feature to the first feature

has_ivls_enclosing(read: velocyto.read.Read) → bool

Finds out if there are intervals that are fully containing all the read segments

Parameters

read (vcy.Read) – the read object to be analyzed

Returns

respones – if one has been found

Return type

bool

mark_overlapping_ivls(read: velocyto.read.Read) → None

Finds the overlap between Read and Features and mark intronic features if spanned

Parameters

read (vcy.Read) – the read object to be analyzed

Returns

Return type

Nothing, it marks the vcy.Feature object (is_validated = True) if there is evidence of exon-intron spanning

find_overlapping_ivls(read: velocyto.read.Read) → Dict[velocyto.transcript_model.TranscriptModel, List[velocyto.segment_match.SegmentMatch]]

Finds the possible overlaps between Read and Features and return a 1 read derived mapping record

Parameters

read (vcy.Read) – the read object to be analyzed

Returns

mapping_record – A record of the mappings by transcript model. Every entry contains a list of segment matches that in turn contains information on the segment and the feature

Return type

Dict[vcy.TranscriptModel, List[vcy.SegmentMatch]]

Note

• It is possible that a segment overalps at the same time an exon and an intron (spanning segment)

• It is not possible that a segment overalps at the same time two exons. In that case the read is splitted

into two segments and the Read attribute is_spliced == True. - Notice that the name of the function might be confousing. if there is a non valid overallapping an empty mappign record will be return - Also notice that returning an empty mapping record will cause the suppression of the counting of the molecule

velocyto.molitem module

velocyto.molitem.dictionary_union(d1: DefaultDict[Any, List], d2: DefaultDict[Any, List]) → DefaultDict[Any, List]

Set union (|) operation on default dicitonary

Parameters

• d1 (defaultdict) – First default dict

• d2 (defaultdict) – Second default dict

Returns

• A dictionary with the key the set union of the keys.

• If same key is present the entry will be combined using __add__

velocyto.molitem.dictionary_intersect(d1: DefaultDict[Any, List], d2: DefaultDict[Any, List]) → DefaultDict[Any, List]

Set intersection (&) operation on default dicitonary

Parameters

• d1 (defaultdict) – First default dict

• d2 (defaultdict) – Second default dict

Returns

• A dictionary with the key the set intersection of the keys.

• If same key is present the entry will be combined using __add__

class velocyto.molitem.Molitem

Bases: object

Object that represents a molecule in the counting pipeline

mappings_record

add_mappings_record(mappings_record: DefaultDict[velocyto.transcript_model.TranscriptModel, List[velocyto.segment_match.SegmentMatch]]) → None

velocyto.gene_info module

class velocyto.gene_info.GeneInfo(genename: str, geneid: str, chromstrand: str, start: int, end: int)

Bases: object

A simple objects that stores basic info on a gene. Parsed from the .gtf file and used to build the row_attrs of the loom file

genename

geneid

chrom

strand

start

end

velocyto.read module

class velocyto.read.Read(bc: str, umi: str, chrom: str, strand: str, pos: int, segments: List, clip5: Any, clip3: Any, ref_skipped: bool)

Bases: object

Container for reads from sam alignment file

bc

umi

chrom

strand

pos

segments

clip5

clip3

ref_skipped

property is_spliced: bool

property start: int

property end: int

property span: int