Input Marker file

[alslonik]

I am trying to construct ancestral chromosomes using ANGeS for 8 plant genomes, and having problems with the input marker file. I have Orthofinder output as well as MCscan and trying to combine the outputs to get the Marker file as shown in the examples of the software. The problem is the the collinearity blocks (from MCscan) are for pairs of genomes and not for all the 8 genomes, and preparing the blocks based on Orthologs (from Orthofinder output) makes the blocks too short for ANGeS. Is there any software that makes it easier to prepare the input marker file? Thank you

[cchauve]

I am not aware of any published method for what you are facing. Generally turning pairwise markers into multi-species markers is challenging and is something AnGeS was not designed for. But I am surprised to read the markers are too short: there is no limit of marker size in AnGeS as far as I remember.

[alslonik]

Thank you for a quick response! I just assumed that the marker being just one gene and not a longer block (as in the software examples) is too short, as i got an error message. Maybe the reason for the error is different.. What we did is we took orthogroups (from Orthofinder software) w/o inparalogs, where each of the species has one ortholog only, and made a marker file of those groups. Does that make sense? Is it enough for AnGEs to recreate the ancestral chromosomes? These are orthologous sequences, not synteny blocks.

Thanks, Alex ---------- Forwarded message --------- From: Computational Methods for Comparative Genomics at Simon Fraser University @.> Date: Wed, 18 Aug 2021, 18:01 Subject: Re: [cchauve/ANGeS] Input Marker file (#2) To: cchauve/ANGeS @.> Cc: alslonik @.>, Author @.>

I am not aware of any published method for what you are facing. Generally turning pairwise markers into multi-species markers is challenging and is something AnGeS was not designed for. But I am surprised to read the markers are too short: there is no limit of marker size in AnGeS as far as I remember.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/cchauve/ANGeS/issues/2#issuecomment-901189295, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJFOT353KOQ7O6ZAAX6YZV3T5PDNPANCNFSM5CLQGXJA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email .

[cchauve]

What is the error message?

[alslonik]

Hi,

I attach the parameters and the input files, as well as the error for your inspection. Thank you!!!

Alex

python /usr/local/ANGeS/src/MASTER/anges_CAR.py PARAMETERS

Parameters File

Project: BOREOEUTHERIAN

Date: 2012-05-03

#

Input files

# markers_file MARKERS_EIGHT_SP.txt # file name for markers tree_file 8sp_timetree_tree.tre # file name for species tree #

Output options

# output_directory ROSIDS # directory name output_ancestor ARK # ancestor name #

Markers options

# markers_doubled 0 # 0 = original markers, 1 = doubled markers markers_unique 0 # 0 = repeated markers allowed, 1 = ingroup unique, 2 = unique markers_universal 1 # 0 = missing markers allowed, 1 = ingroup universal, 2 = universal #

ACS options

#

acs_pairs # name of file containing the species pairs to compare

acs_sa 1 # supported adjacencies: 0 = not computed, 1 = computed acs_ra 0 # reliable adjacencies: 0 = not computed, 1 = computed acs_sci 0 # strong common intervals: 0 = not computed, 1 = computed acs_mci 0 # maximal common intervals: 0 = not computed, 1 = computed acs_aci 0 # all common intervals: 0 = not computed, 1 = computed

acs_file # ACS provided by user

acs_weight 1 # weighting ACS: 1 = linear interpolation acs_correction 2 # Correcting for missing markers: 0 = none, 1 = adding markers spanned by intervals, 2 = X, 3 = both #

C1P model

# c1p_linear 0 # 1 for working with linear chromosomes c1p_circular 0 # 1 for working with a unique circular chromosomes #

Telomeres (model+optimization)

# c1p_telomeres 0 # 0: no telomere, 1: added after optimization (greedy heuristic), 2: added after optimization (bab), 3: added during optimization (bab) #

C1P optimization options

# c1p_heuristic 0 # Using a greedy heuristic c1p_bab 0 # Using a branch-and-bound c1p_spectral 1 # Using spectral seriation #

Spectral seriation options

שגיאה

c1p_spectral_alpha 1.0 # Spectral seriation alpha value #

END

Error message:

----> Started ----> Copying input ----> Filtering markers ----> Markers: ROSIDS/MARKERS/ARK_MARKERS ----> Cleaning ACS file ----> Computing species ------> All species: ROSIDS/ACS/ARK_SPECIES ----> Computing species pairs to compare ------> All informative pairs: ROSIDS/ACS/ARK_PAIRS ----> Computing ancestral contiguous sets: ROSIDS/ACS/ARK_ACS ----> Weighting ACS: ROSIDS/WEIGHT/ARK_WACS ----> Computing PQ-tree from correlation matrix of a ternary matrix using parameter alpha Traceback (most recent call last): File "/usr/local/ANGeS/src/SERIATION/SERIATION_compute_PQtree_dotproduct_correlation_matrix.py", line 92, in for t in T: TypeError: 'NoneType' object is not iterable

On Wed, 18 Aug 2021 at 20:37, Computational Methods for Comparative Genomics at Simon Fraser University @.***> wrote:

What is the error message?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/cchauve/ANGeS/issues/2#issuecomment-901302660, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJFOT37SYSO6UIFOQ6JQNEDT5PVWVANCNFSM5CLQGXJA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email .

--

Kind regards,

Alexandra Slonimsky