initialize matrix in gotoh algorithm */

Using the Kabsch algorithm with two sets of paired point P and Q, centered around the centroid. Each vector set is represented as an NxD

:param x: [bs * n_node, input_size] :param edge_index: list of [n_edge], [n_edge] :param edge_attr: [n_edge, hidden_size], re

:param h: [bs * n_node, input_size] :param edge_index: list of [n_edge], [n_edge] :param att_weight: [n_edge, 1], unsqueezed

Number a sequence and yield the annotations for each domain identified (n to c)

Apply the Aho numbering scheme Rules should be implemented using two strings - the state string and the region string. There a

Apply the Chothia numbering scheme for heavy chains Rules should be implemented using two strings - the state string and the region stri

Apply the IMGT numbering scheme for heavy or light chains Rules should be implemented using two strings - the state string and the r

Apply the Kabat numbering scheme for heavy chains Rules should be implemented using two strings - the state string and the region string

Apply the Kabat numbering scheme for light chains Rules should be implemented using two strings - the state string and the region string

Apply the Martin (extended Chothia) numbering scheme for heavy chains Rules should be implemented using two strings - the state string a

Apply the Martin numbering scheme for light chains Rules should be implemented using two strings - the state string and the region strin

Apply the numbering to each PDB chain in the file

Given you have an alignment. Give back the numbering @param state_vector: List of states from the hmm. Effectively these are imgt column

Given a list of sequences and a corresponding list of alignments from run_hmmer apply a numbering scheme.

Apply the wolfguy numbering scheme for heavy chains The scheme numbers the sequence using different segments so that the numbering tells yo

Apply the wolfguy numbering scheme for light chains The scheme numbers the sequence using different segments so that the numbering tells yo

extract rotation matrix based on TMscore8 */

Interface with pairwise muscle between two sequences that should be identical. Try an easy alignment first by checking that one is in the o

Parse the output of HMMscan and return top alignment and the score table for each input sequence.

Load data from file_path and add processed data entries to self.data. Remember to call self._save_data(num_entry_per_file) to control

Description: Project vector `v` onto vector `e`. Args: v: (N, L, 3). e: (N, L, 3).

Read a sequence file and parse as description, string

Read the seqres entries of a structure if available

Rename the first paired receptor chains with (H, L), (B, A). If single domain then only that is renamed. Deletes all other chains from each m

Find the closest sequence identity match.

side_chain_coord: relative positions in the local coordinate with (CB-CA, N-CA, C-CA) (after gram schmidt)

The function aims to correct to the expected imgt alignment. Renumbering functions then translate from the imgt scheme to the appropriate sch

Split chains with two variable domains into two chains. The first has the original identifier. The second has the lower case identifier.

trace back dynamic programming path to diciper pairwise alignment */

trace back Smith-Waterman dynamic programming path to diciper * pairwise local alignment */

A function to uniquify a sequence preserving order With thanks to http://www.peterbe.com @param seq: A sequence to uniquify @param i

:param data: [n_edge, *dimensions] :param segment_ids: [n_edge] :param num_segments: [bs * n_node]

Wrapper to do some basic validation of the numbering. Further validation could be done but at the moment we just check that the numberin

Write a list of sequences to file. should be a list of name, sequence tuples f should be an open file

Change the numbering scheme used in the identifiers of the residues in the chain. By default the numbering will be with respect to the first