diff --git a/bakta/features/annotation.py b/bakta/features/annotation.py
index 28e1a24..7358aaf 100644
--- a/bakta/features/annotation.py
+++ b/bakta/features/annotation.py
@@ -19,8 +19,12 @@
 RE_PROTEIN_NODE = re.compile(r'NODE_', flags=re.IGNORECASE)
 RE_PROTEIN_POTENTIAL_CONTIG_NAME = re.compile(r'(genome|shotgun)', flags=re.IGNORECASE)
 RE_PROTEIN_DOMAIN_CONTAINING = re.compile(r'domain-containing protein', flags=re.IGNORECASE)
+RE_PROTEIN_REMNANT = re.compile(r'Remnant of ', re.IGNORECASE)
+RE_PROTEIN_TMRNA = re.compile(r'TmRNA', flags=re.IGNORECASE)
 RE_PROTEIN_NO_LETTERS = re.compile(r'[^A-Za-z]')
-RE_PROTEIN_SUSPECT_CHARS = re.compile(r'[.@=?%]')
+RE_PROTEIN_SUSPECT_CHARS_DISCARD = re.compile(r'[.#]')
+RE_PROTEIN_SUSPECT_CHARS_REPLACE = re.compile(r'[@=?%]')
+RE_PROTEIN_SUSPECT_CHARS_BEGINNING = '_\-+.:,;/\\\''
 RE_PROTEIN_PERIOD_SEPARATOR = re.compile(r'([a-zA-Z0-9]+)\.([a-zA-Z0-9]+)')
 RE_PROTEIN_WRONG_PRIMES = re.compile(r'[\u2032\u0060\u00B4]')  # prime (′), grave accent (`), acute accent (´)
 RE_PROTEIN_WEIGHT = re.compile(r' [0-9]+(?:\.[0-9]+)? k?da ', flags=re.IGNORECASE)
@@ -536,9 +540,19 @@ def revise_cds_product(product: str):
     product = re.sub(RE_PROTEIN_PERIOD_SEPARATOR, r'\1-\2', product)  # replace separator periods
     if(product != old_product):
         log.info('fix product: replace separator periods. new=%s, old=%s', product, old_product)
+    
+    old_product = product
+    if(product[0] in RE_PROTEIN_SUSPECT_CHARS_BEGINNING):  # remove suspect first character
+        product = product[1:]
+        log.info('fix product: replace invalid first character. new=%s, old=%s', product, old_product)
+
+    old_product = product
+    product = RE_PROTEIN_SUSPECT_CHARS_DISCARD.sub('', product)  # remove suspect characters
+    if(product != old_product):
+        log.info('fix product: replace invalid characters. new=%s, old=%s', product, old_product)
 
     old_product = product
-    product = RE_PROTEIN_SUSPECT_CHARS.sub('', product)  # remove suspect characters
+    product = RE_PROTEIN_SUSPECT_CHARS_REPLACE.sub(' ', product)  # replace suspect characters by single whitespace
     if(product != old_product):
         log.info('fix product: replace invalid characters. new=%s, old=%s', product, old_product)
 
@@ -552,6 +566,11 @@ def revise_cds_product(product: str):
     if(product != old_product):
         log.info('fix product: replace FOG ids. new=%s, old=%s', product, old_product)
 
+    old_product = product
+    product = RE_PROTEIN_REMNANT.sub('', product)  # remove 'Remnant of's
+    if(product != old_product):
+        log.info('fix product: replace remnant ofs. new=%s, old=%s', product, old_product)
+
     old_product = product
     dufs = []  # replace DUF-containing products
     for m in RE_DOMAIN_OF_UNKNOWN_FUNCTION.finditer(product):
@@ -593,6 +612,11 @@ def revise_cds_product(product: str):
         product = product.replace('_', '-')
         if(product != old_product):
             log.info('fix product: replace domain name underscores. new=%s, old=%s', product, old_product)
+    
+    old_product = product
+    if(RE_PROTEIN_TMRNA.fullmatch(product)):
+        product = ''
+        log.info('fix product: discard pure tmRNA product descriptions. new=%s, old=%s', product, old_product)
 
     old_product = product
     if(