This constraint is open source. Any third-party models, product names, or trademarks referenced are the property of their respective owners, and Proto is not affiliated with them.
evo-design/proto-language/proto_language/constraint/protein_quality/protein_repetitiveness_constraint.pyConstraint contributors
dguo8412
Evaluate protein sequence repetitiveness based on k-mer frequency analysis.
This constraint function analyzes protein sequences for repetitive content by
examining k-mer frequencies. It identifies sequences with excessive repetitive motifs,
which may indicate low-complexity regions or non-functional proteins. The analysis
scans multiple k-mer lengths to detect both short tandem repeats and larger sequence
duplications. The repetitiveness score represents the maximum fraction of the sequence
covered by any repeated k-mer. For example, if “SSS” appears 8 times in a
60-amino-acid sequence, the repetitiveness for 3-mers is (8 * 3) / 60 = 0.4
(40% of sequence).
dguo8412API Reference
Configuration for protein repetitiveness constraint.This class defines configuration parameters for evaluating repetitive content
in protein sequences using k-mer frequency analysis. The constraint detects
and penalizes sequences with excessive tandem repeats or repetitive motifs,
which may indicate low-complexity regions or non-functional proteins. The
repetitiveness score is calculated as the maximum fraction of the sequence
covered by any repeated k-mer. For example, if “AAA” appears 10 times in a
100-amino-acid sequence, the repetitiveness for 3-mers is (10 * 3) / 100 = 0.3
(30% of sequence).
number
default:"0.1"
Maximum acceptable repetitiveness fraction (fraction of sequence covered by repeated k-mers)
integer
default:"1"
Smallest k-mer length treated as a repeat; the scan continues up to this length plus 6.
ReturnsConstraintOutput
One result per sequence. A score of 0.0 indicates
acceptable repetitiveness (at or below threshold) and higher values
indicate excessive repetitive content. Penalties scale linearly with
excess repetitiveness: if max is 0.4 and actual is 0.6, the excess
(0.2) is normalized by the remaining range (1.0 - 0.4 = 0.6), giving
a score of 0.33. metadata carries:repetitiveness_score: Float repetitiveness score (0.0-1.0) representing the maximum fraction of sequence covered by repeated k-mersmax_repetitive_fraction: Float identical torepetitiveness_score
Usage
Evaluating repetitiveness with default settings:python