Repository navigation
Feature/mechanism lookup - #27
Merged
Merged
Conversation
States are stored in the order Mechanism sorts them into, by state type:
open first, then short-lived shut, then long-lived shut. That is rarely the
order a paper tabulates them in. CH82 is stored as AR*, A2R*, AR, A2R, R,
and the ten-state flip mechanism of Burzomato et al. (2004) differs from the
published ordering by a swap of two states. Reading a Q matrix by position
against a published table therefore invites an error that nothing catches,
because the wrong entry is still a plausible number.
Adds, all computed on demand and none of them caching anything:
state_names names, in Q-matrix order
state_index(name) row and column of a state
rates_by_edge() {(from, to): Rate} for every transition
rate(from, to) one transition, KeyError if there is no such step
permutation(order) index array to reorder Q into a given order
rate() raising rather than returning None is the point of it: asking for a
step the mechanism does not have is a question about a different mechanism.
When a name is simply unknown it says so, and lists the states there are.
Reproducing a published paper previously meant writing a wrapper class to do
this; that is what prompted it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
transition_probability had three implementations, two in scalcslib and one
in scsim; transition_frequency had two. They had drifted:
- scalcslib's copies carried empty docstrings;
- scalcslib's transition_probability did Q.copy() without casting, so an
integer Q matrix divided integer-wise and returned silently wrong
numbers rather than raising;
- scalcslib's transition_frequency computed pinf internally and so could
only ever give equilibrium frequencies.
Both now live in qmatlib, with the rest of the plain Q-matrix algebra, and
are re-exported from scalcslib and scsim so nothing importing them breaks.
transition_frequency now takes the occupancies as an optional argument.
Called with one argument it still uses pinf, which is what every existing
call site does; passing p gives the frequencies at an instant during a
concentration jump, when the system is not at equilibrium.
The tests pin the definitions, that neither function modifies its argument,
that an integer Q is no longer truncated, and that at equilibrium detailed
balance holds edge by edge. That last one is only approximate for CH82,
whose MR-constrained rate is stored rounded to 0.66667, so its cycle closes
to about 1e-9; a mechanism whose rates come from a fit closes to 1e-12.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
update_mr() imposes the constraint. Nothing reported whether it actually holds for the rate constants a mechanism is carrying, which is a different question and a useful one: a set that has been averaged, transcribed, or written by another program can fail it, and the size of the failure says how much to trust the set. cycle_ratios() [(states, forward/backward), ...], one per cycle reversibility_error() max |ratio - 1| over all cycles, 0.0 if none check_mr already returned the two products for one cycle, and these agree with it - there is a test pinning that so they cannot drift. What they add is the whole-mechanism summary, and failing loudly: check_mr looks each step up with a scan that contributes nothing when no rate matches, so a cycle naming a transition the mechanism does not have comes back as a perfectly closed cycle. These use rate(), which raises. Scale, for reading the number. The two ten-state datasets behind Colquhoun & Lape (2012): single fitted set 1.4e-07 (stored as 32-bit floats) averaged set 4.2e-02 (ratios 0.9576 and 0.9671) Averaging several fitted sets does not preserve cycle conditions, and that is what a violation of percents rather than parts in 10^7 looks like. CH82 gives 5e-06, because its constrained rate is stored rounded to 0.66667. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Merged
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.